Laguerre reconstruction of the BAO feature in halo-based mock galaxy catalogues

TL;DR

This paper demonstrates that Laguerre function fitting effectively reconstructs the initial BAO feature in various mock galaxy catalogs, providing a bias-insensitive and precise distance scale measurement method.

Contribution

The study introduces a Laguerre reconstruction method that accurately recovers the BAO feature in complex galaxy mocks, simplifying bias marginalization and improving distance scale estimates.

Findings

Laguerre reconstruction reduces bias-related offsets to sub-percent levels.

The method is insensitive to galaxy bias at the sub-percent level.

Laguerre reconstruction simplifies parameter marginalization without extra computational cost.

Abstract

Fitting half-integer generalized Laguerre functions to the evolved, real-space dark matter and halo correlation functions provides a simple way to reconstruct their initial shapes. We show that this methodology also works well in a wide variety of realistic, assembly biased, velocity biased and redshift-space distorted mock galaxy catalogs. We use the linear point feature in the monopole of the redshift-space distorted correlation function to quantify the accuracy of our approach. We find that the linear point estimated from the mock galaxy catalogs is insensitive to the details of the biasing scheme at the sub-percent level. However, the linear point scale in the nonlinear, biased, and redshift-space distorted field is systematically offset from its scale in the unbiased linear density fluctuation field by more than 1%. In the Laguerre reconstructed correlation function, this is reduced to sub-percent values, so it provides comparable accuracy and precision to methods that reconstruct the full density field before estimating the distance scale. The linear point in the reconstructed density fields provided by these other methods is likewise precise, accurate, and insensitive to galaxy bias. All reconstructions depend on some input parameters, and marginalizing over uncertainties in the input parameters required for reconstruction can degrade both accuracy and precision. The linear point simplifies the marginalization process, enabling more realistic estimates of the precision of the distance scale estimate for negligible additional computational cost. We show this explicitly for Laguerre reconstruction.