Tidal alignments as a contaminant of the galaxy bispectrum

TL;DR

Galaxy intrinsic alignments, if correlated with large-scale structure, can contaminate the galaxy bispectrum measurements, affecting cosmological parameter estimates, but some effects can be mitigated with survey strategies.

Contribution

This paper models the impact of galaxy intrinsic alignments on the galaxy bispectrum and explores how these effects influence cosmological parameter inference.

Findings

Linear alignment model affects galaxy bias estimates but not sigma_8.

Quadratic alignment model causes systematic errors in bias and sigma_8.

Configuration dependence may allow removal of quadratic alignment contamination.

Abstract

If the orientations of galaxies are correlated with large-scale structure, then anisotropic selection effects such as preferential selection of face-on disc galaxies can contaminate large scale structure observables. Here we consider the effect on the galaxy bispectrum, which has attracted interest as a way to break the degeneracy between galaxy bias and the amplitude of matter fluctuations sigma_8. We consider two models of intrinsic galaxy alignments: one where the probability distribution for the galaxy's orientation contains a term linear in the local tidal field, appropriate for elliptical galaxies; and one with a term quadratic in the local tidal field, which may be applicable to disc galaxies. We compute the correction to the redshift-space bispectrum in the quasilinear regime, and then focus on its effects on parameter constraints from the transverse bispectrum, i.e. using triangles in the plane of the sky. We show that in the linear alignment model, intrinsic alignments result in an error in the galaxy bias parameters, but do not affect the inferred value of sigma_8. In contrast, the quadratic alignment model results in a systematic error in both the bias parameters and sigma_8. However, the quadratic alignment effect has a unique configuration dependence that should enable it to be removed in upcoming surveys.