Abstract

Gravitational waves opened a new window to study the properties of compact binary systems, providing insights into their astrophysical formation and evolution. Initially, population studies relied on mass and spin distributions to distinguish between isolated, dynamical, and hierarchical merger channels; however, orbital eccentricity offers a complementary probe of binary formation scenarios. Historically, eccentricity received limited attention due to the lack of eccentric wave- form models and the absence of strong observational evidence within the LIGO–Virgo–KAGRA sensitivity band. A growing catalog, recent advances in eccentric waveform modeling, and analyses reporting candidates consistent with non-zero eccentricity motivate a systematic population-level study with eccentricity. This requires a computationally efficient framework to extract meaningful population-level information using mixture models. In this work, we develop a JAX-based population inference framework, GWKokab, designed to construct complex population models from simple, modular components through a command-line interface. The framework is validated by reproducing previously published results at significantly reduced computational cost. Since spin and eccentricity often occupy narrow regions of parameter space and peak near zero, GWKokab also has an analytic population likelihood to complement standard discrete methods and mitigate biases arising in edge-dominated regimes. For potential science studies, GWKokab also offers the capability to generate spinning-eccentric synthetic catalogs and corresponding posterior samples, including real posteriors obtained with eccentric waveform models using RIFT. Our population inference results are based on parameter estimation performed with the ec- centric waveform models SEOBNRv5EHM and TEOBResumSDALI on the GWTC-4 catalog. We compare the population properties under quasi-circular and eccentric assumptions using the SEOBNRv5EHM parameter estimation results. In this study, we also introduce a mixture of half-Normal and truncated-Normal distributions for the eccentricity distribution. We found that the overall population properties are consistent with the quasi-circular assumption and that only ∼2% of binaries may be eccentric in the current catalog. To extend this work further, we study waveform systematics using the two models and find that waveform systematics can produce biases at the population level even when individual events agree well, particularly for the redshift and effective-spin distributions. Lastly, to probe hierarchical formation in the GWTC-5 catalog of binary black holes, we introduce a multi-component parametric population model consisting of one power-law and five Gaussian components with independent rates and spins. Our multi-component model shows a roughly hierarchical spectrum of Gaussian mass peaks, but without the expected correlations between spin and mass predicted by naked hierarchical formation.

Publication Date

8-2026

Document Type

Dissertation

Student Type

Graduate

Degree Name

Astrophysical Sciences and Technology (Ph.D.)

Department, Program, or Center

Physics and Astronomy, School of

College

College of Science

Advisor

Richard O’Shaughnessy

Advisor/Committee Member

George Thurston

Advisor/Committee Member

Thomas Callister

Campus

RIT – Main Campus

Download

Share

COinS