Current and future constraints on cosmology and modified gravitational wave friction from binary black holes

TL;DR

This paper investigates how gravitational wave observations from binary black hole mergers can constrain modified gravity theories, focusing on the effects of an extra friction term on luminosity distances and forecasting future measurement precisions.

Contribution

It introduces a joint analysis method for BBH population, cosmological, and modified gravity parameters using LIGO/Virgo data, and forecasts future constraints on modified gravity parameters.

Findings

General relativity remains the preferred model based on current data.

Forecasted measurement uncertainty on $\\Xi_0$ is 51% with O4 and 20% with O4 and O5.

Strong correlations between parameters may bias results if priors are incorrect.

Abstract

In this proceedings, we are interested in dark gravitational wave standard sirens and their use for cosmology and for constraining modified gravity theories. Due to the extra friction term introduced in their propagation equation those theories predict different luminosity distances for electromagnetic and gravitational waves (GWs). This effect can be parametrized by the two variables $\Xi_0$ and $n$, that can be measured from gravitational wave observations, and specifically from the binary black hole (BBH) mergers detected by LIGO and Virgo. By fitting jointly BBH population models in mass and redshift, the cosmological parameters, and the modified GW luminosity distance to $\sim$ 60 signals observed during the first three LIGO/Virgo observation runs, we conclude that general relativity is consistently the preferred model. The future observation runs O4 and O5 are also considered. Using the same approach, we forecast a measurement uncertainty on the modified gravity parameter $\Xi_0$ of $51\%$ with O4, and $20\%$ with O4 and O5, respectively if GR is the correct theory of gravity. However, we underline that there are strong correlations between astrophysical, cosmological and modified gravity parameters, possibly leading to bias if wrong priors are assumed.