Cosmic Cousins: Identification of a Subpopulation of Binary Black Holes Consistent with Isolated Binary Evolution

TL;DR

This paper identifies a subpopulation of binary black holes with properties consistent with isolated binary evolution, using hierarchical Bayesian inference on GWTC-3 data, revealing insights into black hole formation channels.

Contribution

It introduces a novel hierarchical Bayesian method with discrete latent variables to classify black hole mergers into distinct formation channels based on mass and spin correlations.

Findings

Detected a low-mass subpopulation (~10 M_sun) with aligned spins.

Found the 35 M_sun peak's spins are consistent with the low-mass group.

Estimated the lower edge of the pair instability mass gap at ~49 M_sun.

Abstract

Observations of gravitational waves (GWs) from merging compact binaries have become a regular occurrence. The continued advancement of the LIGO-Virgo-KAGRA (LVK) Collaboration detectors have now produced a catalog of over 90 such mergers, from which we can begin to uncover the formation history of merging compact binaries. In this work, we search for subpopulations in the LVK's third gravitational wave transient catalog (GWTC-3) by incorporating discrete latent variables in the hierarchical Bayesian inference framework to probabilistically assign each BBH observation into separate categories associated with distinctly different population distributions. By incorporating formation channel knowledge within the mass and spin correlations found in each category, we find an over density of mergers with a primary mass of $\sim10 M_\odot$, consistent with isolated binary formation. This low-mass subpopulation has a spin magnitude distribution peaking at $a_\mathrm{peak}=0.16^{0.19}_{-0.16}$, exhibits spins preferentially aligned with the binary's orbital angular momentum, is constrained by $15^{+0.0}_{-1.0}$ of our observations, and contributes $82\%^{+8.0\%}_{-16\%}$ to the overall population of BBHs. Additionally, we find that the component of the mass distribution containing the previously identified $35M_\odot$ peak has spins consistent with the $10M_\odot$ events, with $99\%$ of primary masses less than $m_{1,99\%} = 49^{+25}_{-8.1} M_\odot$, providing an estimate of the lower edge of the theorized pair instability mass gap. This work is a first step in gaining a deeper understanding of compact binary formation and evolution, and will provide more robust conclusions as the catalog of observations becomes larger.