Second-order matter fluctuations via higher-order galaxy correlators

TL;DR

This paper introduces a new method to measure the amplitude of matter fluctuations directly from galaxy survey data, independent of bias modeling, enabling tests of gravity on large scales.

Contribution

It presents a bias-insensitive estimator for sigma_R using galaxy correlators, applicable to large-scale surveys, and validates it with N-body simulations.

Findings

Successfully recovered sigma_8 in simulations

Estimator is insensitive to bias function details

Provides a consistency check for real data analysis

Abstract

We provide a formula for extracting the value of the rms of the linear matter fluctuations on a scale R directly from redshift surveys data. It allows to constrain the real-space amplitude of sigma_R without requiring any modeling of the nature and power spectrum of the matter distribution. Furthermore, the formalism is completely insensitive to the character of the bias function, namely its eventual scale or non-linear dependence. By contrasting measurements of sigma_R with predictions from linear perturbation theory, one can test for eventual departures from the standard description of gravity on large cosmological scales. The proposed estimator exploits the information contained in the 1-point moments and 2-point correlators of the matter and galaxy density fields, and it can be applied on cosmic scales where linear and semi-linear perturbative approximations of the evolution of matter overdensities offer a satisfactory description of the full underlying theory. We implement the test with N-body simulations to quantify potential systematics and successfully show that we are able to recover the present day value of sigma_8 `hidden' in the simulation. We also design a consistency check to gauge the soundness of the results inferred when the formalism is applied to real (as opposed to simulated) data. We expect that this approach will provide a sensitive probe of the clustering of matter when applied to future large redshift survey such as BigBOSS and EUCLID.