Self-consistent predictions of electromagnetic and gravitational wave stellar transients

Abstract

Since 2015 gravitational-wave (GW) observations have provided valuable insight into the properties of the merging binary black hole (BBH) population. The most striking features of the observed primary black hole mass distributions are the extended tail up to 100M⊙ and an excess of masses at 35M⊙. However, traditional isolated binary population synthesis has difficulty explaining these two features. To address this uncertainty, we have created a synthetic population of stellar transients using the detailed binary population synthesis, BPASS, and star formation histories (SFH) based on observations and cosmological simulations. This self-consistent approach allows for the prediction of multiple observables from the same population, providing robust constraints on the implemented stellar evolution. We are able to match supernova (SN) and GW observations simultaneously. Furthermore, various types of transients, including Type Ib/c and pair-instability SNe, probe the metallicity of star formation, while Type Ia SNe probes older star formation regions. The comprehensive matching of observations from a single cosmic population further validifies our approach. In addition to matching the SN observations, the synthetic population reproduces the observed BBH mass distribution features, although not through the traditionally expected mechanisms of pulsational pair-instability. Instead, we identify the stability of mass transfer, quasi-homogenous evolution, and stellar winds as essential processes to reproduce the 35M⊙ excess, whilst the extended tail results from super-Eddington accretion during stable mass transfer at high mass ratios onto a black hole. These key features are independent of the SFH and remnant mass prescription, including (pulsational) pairinstability SNe. However, the 35M⊙ peak is redshift dependent and disappears at high redshift due to the delay time distribution of the formation channel. Since the mass transfer stability allows for both features to occur, we explore this in detail and show that other binary population synthesis codes do not accurately capture it, thus missing the features in their populations. Future population synthesis codes should include detailed prescriptions for mass transfer and make predictions for multiple observables to constrain their implemented physics.