Quantifying effects of inhomogeneities and curvature on gravitational wave standard siren measurements of $H(z)$

TL;DR

This paper investigates how inhomogeneities and curvature affect gravitational wave-based measurements of the universe's expansion rate, revealing challenges in detecting cosmic backreaction and curvature effects with current and future data.

Contribution

It demonstrates that curvature and backreaction can bias H(z) measurements from dipole luminosity distances, and assesses the detectability of these effects with mock gravitational wave data.

Findings

1% curvature may be barely detectable in ideal conditions.

5% curvature leads to unreliable estimates of inconsistency.

Backreaction signals are difficult to detect unless they are very large.

Abstract

For a flat $\Lambda$CDM universe, the dipole of the luminosity distance can be utilized to measure the Hubble parameter. It is here shown that this is not the case in more general settings where curvature and cosmic backreaction is permitted. This implies that a discordance between $H(z)$ measurements obtained using such dipole luminosity distance data and "true"/actual $H(z)$ data obtained from e.g. cosmic chronometers is a signal of curvature and/or cosmic backreaction. \newline\indent By considering mock future gravitational wave measurements of the Hubble parameter obtained through the dipole luminosity distance, it is shown that already a $1\%$ curvature could in principle just barely show up in the determination. However, for realistic mock data generation using models with as much as 5 \% curvature, parameter estimates do not yield reliable measures of inconsistency between the false $H(z)$ measurements and true measurements of $H(z)$. At the same time, cosmic backreaction is hard to detect even if it makes up $10\%$ of the "energy budget" in the current universe, even when considering a highly idealized situation with low errors. The results concerning backreaction are based on specific "scaling solutions" to the backreaction problem and the study shows that the possibility of detecting a signal of backreaction through the dipole of the luminosity distance depends strongly on the particular backreaction model.