Baryon oscillations in galaxy and matter power-spectrum covariance matrices

TL;DR

This paper analyzes baryon acoustic oscillations in the covariance matrices of galaxy and matter power spectra, deriving an analytic approximation and assessing their detectability and impact on cosmological parameter estimation.

Contribution

It introduces an analytic model for BAO features in covariance matrices and evaluates their significance in galaxy surveys and parameter constraints.

Findings

BAO features in covariance matrices are stronger than in power spectra.

These covariance BAO's are detectable at about one-sigma in large surveys.

Including covariance BAO's affects error estimates for certain cosmological parameters.

Abstract

We investigate large-amplitude baryon acoustic oscillations (BAO's) in off-diagonal entries of cosmological power-spectrum covariance matrices. These covariance-matrix BAO's describe the increased attenuation of power-spectrum BAO's caused by upward fluctuations in large-scale power. We derive an analytic approximation to covariance-matrix entries in the BAO regime, and check the analytical predictions using N-body simulations. These BAO's look much stronger than the BAO's in the power spectrum, but seem detectable only at about a one-sigma level in gigaparsec-scale galaxy surveys. In estimating cosmological parameters using matter or galaxy power spectra, including the covariance-matrix BAO's can have a several-percent effect on error-bar widths for some parameters directly related to the BAO's, such as the baryon fraction. Also, we find that including the numerous galaxies in small haloes in a survey can reduce error bars in these cosmological parameters more than the simple reduction in shot noise might suggest.