Bayesian parameter estimation for the Core-bounce phase of Rapidly Rotating Core-Collapse Supernovae in real interferometric data

TL;DR

This paper introduces a new Bayesian method to estimate core-collapse supernova parameters from gravitational-wave data, improving accuracy and robustness over previous models.

Contribution

It extends a phenomenological waveform model with an additional parameter, enhancing parameter estimation accuracy using real interferometric noise.

Findings

Median fitting factor increased from 88.88% to 90.83%.

Rotational parameter β recovered with median error of 11.93%.

Biases due to noise can be reduced with optimized priors.

Abstract

We present a novel methodology to estimate the ratio of kinetic to gravitational potential energy in core-collapse supernova progenitors and to assess the equation of state (EOS) using gravitational-wave signals from the core-bounce phase of rapidly rotating stars in real interferometric data. We extend a previous phenomenological model by introducing an additional parameter that captures the signal timescale. The agreement between our template and numerical waveform databases is evaluated through fitting factors and Bayesian model comparison, also assessing consistency across datasets. The improved model increases the median fitting factor from 88.88% to 90.83%. Parameter estimation is performed via Markov Chain Monte Carlo using real O3aL1 noise. For 452 simulated signals, the rotational parameter $\beta$ is recovered with a median relative error of 11.93% (95th percentile: 38.41%) and an uncertainty of $\sigma_\beta = 1.083 \times 10^{-3}$ at 10 kpc, improving over previous matched-filtering results. We further analyze the impact of prior choices and noise properties, finding that real interferometric noise introduces biases up to 11.9%, while optimized priors can reduce them to 0.6%.