The large-scale 3-point correlation function of the SDSS BOSS DR12 CMASS galaxies

TL;DR

This paper measures the large-scale 3-point correlation function of galaxies from the SDSS BOSS DR12 dataset, using novel algorithms and models to analyze BAO features and galaxy bias with high precision.

Contribution

It introduces an efficient algorithm for computing the 3PCF multipole moments and applies it to a large galaxy dataset to measure bias and BAO signals.

Findings

Measured galaxy bias with 2.6% precision

Detected a 2.8σ preference for BAO in the 3PCF

Validated perturbation theory models for large-scale structure

Abstract

We report a measurement of the large-scale 3-point correlation function of galaxies using the largest dataset for this purpose to date, 777, 202 Luminous Red Galaxies in the Sloan Digital Sky Survey Baryon Acoustic Oscillation Spectroscopic Survey (SDSS BOSS) DR12 CMASS sample. This work exploits the novel algorithm of Slepian & Eisenstein (2015b) to compute the multipole moments of the 3PCF in $\mathcal{O}(N^2)$ time, with $N$ the number of galaxies. Leading-order perturbation theory models the data well in a compressed basis where one triangle side is integrated out. We also present an accurate and computationally efficient means of estimating the covariance matrix. With these techniques the redshift-space linear and non-linear bias are measured, with 2.6% precision on the former if $\sigma_8$ is fixed. The data also indicates a $2.8\sigma$ preference for the BAO, confirming the presence of BAO in the 3-point function.