Heavy Black-Holes Also Matter in Standard Siren Cosmology

TL;DR

The paper addresses the challenge of constraining the Hubble constant $H_0$ with gravitational-wave standard sirens and investigates how the assumed CBC mass spectrum affects $H_0$ inference. It introduces a new population model that adds a high-mass feature around $63$ $M_odot$ to the existing FullPop-4.0 framework and performs hierarchical Bayesian inference on 142 CBCs from GWTC-4.0 using both spectral and dark sirens. The key finding is that this heavy-mass feature anti-correlates with $H_0$ and, together with refined constraints on lower-mass peaks, yields a ~30–36% improvement in $H_0$ precision for GWTC-4.0 data and ~12.9% when combined with the bright siren GW170817, though the tension between Planck and SH0ES remains unresolved. This work demonstrates that heavy black-hole populations can significantly boost standard-siren cosmology and should be incorporated in future analyses to maximize cosmological leverage.

Abstract

With the release of the Gravitational-Wave Transient Catalog GWTC-4.0 by the LIGO-Virgo-KAGRA (LVK) collaboration, 218 candidate detections of gravitational waves (GWs) from compact binary coalescences (CBCs) have been reported. This milestone represents a major advancement for GW cosmology, as many methods, particularly those employing the spectral siren approach, critically depend on the number of available sources. We investigate the impact of a novel parametric model describing the full population mass spectrum of CBCs on the estimation of the Hubble constant. This model is designed to test the impact of heavy black holes in GW cosmology. We perform a joint inference of cosmological and population parameters using 142 CBCs from GWTC-4.0 with a false alarm rate smaller than 0.25 per year, using both spectral and dark siren approaches. With spectral sirens, we estimate the Hubble constant to be $H_0 = 78.8^{+19.0}_{-15.3}\,{\rm km s^{-1} Mpc^{-1}}$ (68% CL), while the dark siren method yields $H_0 = 82.5^{+16.8}_{-14.3}\,{\rm km s^{-1} Mpc^{-1}}$ (68% CL). These results improve the uncertainty by approximately 30.4% and 36.2%, respectively. We show that this improvement is linked to the presence of a new mass scale in the binary black hole mass spectrum at $63.3^{+4.8}_{-4.8}\,M_{\odot}$, which provides additional constraints on the Hubble constant. Besides providing the tightest standard-siren constraints on $H_0$, this highlights the importance of a heavy-mass feature in the black hole spectrum.

Figures (4)

• Figure 1: Marginalized Hubble constant posteriors for different analyses. The fiducial GWTC-4.0 model with $\textsc{FullPop-4.0}\xspace$ mass model is shown in blue, and this work shown in pink. The solid lines with filled curves present the spectral siren posteriors, with the dark siren posteriors presented with the dashed lines. Vertical lines are the Hubble tension reference values from Planck and SH0ES Planck:2015fieRiess:2021jrx.
• Figure 2: Inferred 1D posteriors and 2D-contours of the Hubble constant $H_0$ and the position of the different mass scales, for both the $\textsc{FullPop-4.0}\xspace$ (blue) and our mass model (pink). The results were obtained with the dark siren inference. The contours shows the $1\sigma$ and $2\sigma$ levels.
• Figure 3: Mass scale evolution estimated from the dark siren inference of GWTC-4.0, visualized in the detector frame $\{d_{\rm L}, m_{\rm det}\}$. The solid lines represent the MAP value of the different mass scales of our population model, namely the minimum and maximum masses (black), the first (green), second (blue) and third (pink) Gaussian peaks. The shaded regions show the $68\%$ C.L.
• Figure 4: Posterior predictive distribution of the primary mass spectrum, with the $\textsc{FullPop-4.0}\xspace$ in blue and our model in pink. The thick lines denote the median shape of the distribution, while the filled regions give the $68\%$ C.L.