[Paper Review] Bayesian model-selection of neutron star equation of state using multi-messenger observations

B. Biswas|arXiv (Cornell University)|Jun 4, 2021

Pulsars and Gravitational Waves Research4 citations

TL;DR

This study performs Bayesian model-selection among 31 nuclear-physics motivated neutron star equation of state (EoS) models using multi-messenger data, including gravitational wave signals from GW170817 and GW190425, and mass-radius measurements from NICER and radio observations of PSR J0030+0451 and PSR J0740+6620. The AP3 (or MPA1) EoS is found to be the most preferred, with a 1.4 M⊙ neutron star radius of 12.10 km and tidal deformability of 393, while extremely stiff or soft EoSs like MS1, MS1B, SKI5, H4, and WFF1 are decisively ruled out.

ABSTRACT

Measurement of macroscopic properties of neutron stars, whether in binary or in an isolated system, provides us a key opportunity to place a stringent constraint on its equation of state. In this paper, we perform Bayesian model-selection on a wide variety of neutron star equation of state using multi-messenger observations. In particular, (i) we use the mass and tidal deformability measurement from two binary neutron star merger event, GW170817 and GW190425; (ii) simultaneous mass-radius measurement of PSR J0030+0451 and PSR J0740+6620 by NICER collaboration, while the latter has been analyzed by joint NICER/radio/XMM-Newton collaboration. Among the 31 equations of state considered in this analysis, we are able to rule out different variants of MS1 family, SKI5, H4, and WFF1 EoSs decisively, which are either extremely stiff or soft equations of state. The most preferred equation of state model turns out to be AP3 (or MPA1), which predicts the radius and dimensionless tidal deformability of a $1.4 M_{\odot}$ neutron star to be 12.10 (12.50) km and 393 (513) respectively.

Motivation & Objective

• To determine the most statistically preferred neutron star equation of state (EoS) among a wide range of nuclear-physics motivated models using current multi-messenger observations.
• To assess the statistical evidence for each EoS model in light of combined gravitational wave and electromagnetic constraints.
• To provide a benchmark for future EoS modeling and numerical simulations by establishing the current status of EoS models based on observational data.
• To evaluate the robustness of EoS selection under different data combinations, including revised mass measurements of PSR J0740+6620.

Proposed method

• Bayesian model-selection is applied to 31 neutron star EoS models derived from various nuclear many-body approximations.
• The analysis combines constraints from gravitational wave observations of binary neutron star mergers GW170817 and GW190425, which provide mass and tidal deformability measurements.
• Electromagnetic constraints from NICER and radio observations are incorporated, specifically the mass-radius measurements of PSR J0030+0451 and PSR J0740+6620.
• The evidence for each EoS model is computed using the Bayesian evidence (marginal likelihood), allowing for model comparison via Bayes factors.
• Macroscopic properties such as radius and tidal deformability for 1.4 M⊙ neutron stars are calculated using the LalSuite package for all EoS models.
• The analysis excludes EoSs with phase transitions except for HQC18, which is included as a hybrid quark-hadron matter model.

Experimental results

Research questions

• RQ1 Which neutron star equation of state is most strongly supported by current multi-messenger observations?
• RQ2 How do gravitational wave and electromagnetic observations jointly constrain the stiffness and softness of neutron star matter?
• RQ3 Which EoS models are decisively ruled out by the combined data, and what are the quantitative limits on radius and tidal deformability?
• RQ4 How does the inclusion of revised mass measurements of PSR J0740+6620 affect the model selection outcome?
• RQ5 To what extent do the results align with predictions from nuclear physics data, such as those from Li & Steiner (2006)?

Key findings

• The AP3 (or MPA1) equation of state is identified as the most preferred model, with a 1.4 M⊙ neutron star radius of 12.10 km and tidal deformability of 393.
• The EoS models MS1, MS1B, MS1B_PP, MS1_PP, and SKI5 are decisively ruled out due to their extreme stiffness, predicting radii ≥14.07 km and tidal deformabilities ≥1009.
• The WFF1 EoS is decisively ruled out due to its extreme softness, predicting a 1.4 M⊙ radius of 10.42 km and tidal deformability of 153.
• The H4 EoS is also decisively ruled out, predicting a radius of 13.69 km and tidal deformability of 897 for a 1.4 M⊙ neutron star.
• The analysis shows that neutron star EoSs cannot be either very stiff or very soft, with the preferred radius range for 1.4 M⊙ stars being consistent with 11.5–13.6 km as predicted by nuclear physics data.
• The methodology enables future integration of new multi-messenger observations through a consistent Bayesian framework for EoS model comparison.