Searching for Modified Gravity: Scale and Redshift Dependent Constraints from Galaxy Peculiar Velocities

TL;DR

This paper tests modifications to gravity using galaxy surveys and cosmological data, finding results consistent with general relativity at 95% confidence, but noting some tension when allowing expansion history deviations.

Contribution

It introduces a scale- and time-dependent parametrization of modified gravity and jointly constrains it with multiple cosmological probes, including galaxy velocities and power spectra.

Findings

Best-fit parameters are consistent with GR at 95% CL.

Measured growth index gamma aligns with GR predictions.

Some tension arises when allowing deviations in expansion history.

Abstract

We present measurements of both scale- and time-dependent deviations from the standard gravitational field equations. These late-time modifications are introduced separately for relativistic and non-relativistic particles, by way of the parameters $G_{\rm matter}(k,z)$ and $G_{\rm light}(k,z)$ using two bins in both scale and time, with transition wavenumber $0.01$ Mpc$^{-1}$ and redshift 1. We emphasize the use of two dynamical probes to constrain this set of parameters, galaxy power spectrum multipoles and the direct peculiar velocity power spectrum, which probe fluctuations on different scales. The multipole measurements are derived from the WiggleZ and BOSS Data Release 11 CMASS galaxy redshift surveys and the velocity power spectrum is measured from the velocity sub-sample of the 6-degree Field Galaxy Survey. We combine with additional cosmological probes including baryon acoustic oscillations, Type Ia SNe, the cosmic microwave background (CMB), lensing of the CMB, and the temperature--galaxy cross-correlation. Using a Markov Chain Monte Carlo likelihood analysis, we find the inferred best-fit parameter values of $G_{\rm matter}(k,z)$ and $G_{\rm light}(k,z)$ to be consistent with the standard model at the $95\%$ confidence level. Furthermore, accounting for the Alcock-Paczynski effect, we perform joint fits for the expansion history and growth index gamma; we measure $\gamma = 0.665 \pm 0.0669$ ($68\%$ C.L) for a fixed expansion history, and $\gamma = 0.73^{+0.08}_{-0.10}$ ($68\%$ C.L) when the expansion history is allowed to deviate from $\Lambda$CDM. With a fixed expansion history the inferred value is consistent with GR at the $95\%$ C.L; alternatively, a $2\sigma$ tension is observed when the expansion history is not fixed, this tension is worsened by the combination of growth and SNe data.