From Multimessenger Inference to Simulations: A Ranked Ensemble of Finite-Temperature Equations of State

Abstract

We construct a set of microphysical, finite-temperature equations of state (EOSs) for numerical simulations of neutron star mergers and core-collapse supernovae that is consistent with modern constraints from nuclear theory and multimessenger astronomy and systematically spans the posterior distribution of allowed EOSs. The EOSs are based on a simplified Skyrme functional whose inputs are nuclear matter saturation properties, extended to supra-saturation densities through the speed of sound at a set of reference densities. The models are assigned continuous likelihood weights from chiral effective field theory and perturbative quantum chromodynamics calculations, the gravitational wave signal GW170817, the complete set of NICER mass-radius measurements, and the Shapiro-delay mass measurement of the radio pulsar J0348. From the resulting catalogue of 5.2 million EOSs, the posterior yields $R_{1.4} = 11.8^{+0.8}_{-0.7}$ km, $Λ_{1.4} = 334^{+193}_{-113}$, and $M_{\rm max}^{\rm TOV} = 2.18^{+0.22}_{-0.14}\,M_\odot$ (medians with 90% credible intervals). From this posterior we select a ranked 12-member ensemble, headed by a fiducial, central EOS, whose members are individually plausible while jointly bracketing the posterior spread of neutron-star observables. For all ensemble members we generate general-purpose finite-temperature tables with the SROEOS code, each accompanied by nucleon effective-mass variants that bracket the ensemble's thermal-sector uncertainty, to be made publicly available upon publication.