Prior Probability Distributions of Neutron Star Crust Models

To make best use of multifaceted astronomical and nuclear data sets, probability distributions of neutron star models that can be used to propagate errors consistently from one domain to another are required. We take steps toward a consistent model for this purpose, highlight where model inconsistencies occur, and assess the resulting model uncertainty. Using two distributions of nuclear symmetry energy parameters-one uniform, the other based on pure neutron matter theory-we prepare two ensembles of neutron star inner crust models. We use an extended Skyrme energy density functional within a compressible liquid drop model (CLDM). We fit the surface parameters of the CLDM to quantum 3D Hartree-Fock calculations of crustal nuclei. All models predict that more than 50% of the crust by mass and 15% of the crust by thickness comprises pasta with medians of around 62% and 30%, respectively. We also present 68% and 95% ranges for the crust composition as a function of density. We examine the relationships between crust-core boundary and pasta transition properties, the thickness of the pasta layers, the symmetry energy at saturation and subsaturation densities, and the neutron skins of Pb-208 and Ca-48. We quantify the correlations using the maximal information coefficient, which can effectively characterize nonlinear relationships. Future measurements of neutron skins, information from nuclear masses and giant resonances, and theoretical constraints on PNM will be able to place constraints on the location of the pasta and crust-core boundaries and the amount of pasta in the crust.