Estimating nucleon substructure properties in a unified model of p-Pb and Pb-Pb collisions

Abstract

We apply a well tested hybrid transport model, which couples viscous hydrodynamics to a hadronic afterburner, to describe bulk observables in proton-lead and lead-lead collisions at $\sqrt{s_{NN}}=5.02$ TeV. The quark-gluon plasma (QGP) initial conditions are modeled using the parametric TRENTO model with additional nucleon substructure parameters to vary the number and size of hot spots inside each nucleon, followed by a pre-equilibrium free streaming stage to match the full energy-momentum tensor of the initial state onto viscous hydrodynamics. Initial condition and QGP medium parameters, such as the temperature dependence of the QGP shear and bulk viscosities, are then calibrated using Bayesian parameter estimation to describe charged particle yields, mean $p_T$ and anisotropic flow harmonics of both collision systems in a single self-consistent framework. We find that the hybrid model provides a compelling, simultaneous description of both collision systems using appropriately chosen model parameters, and present new posterior estimates for the size and shape of the nucleon and temperature dependence of QGP shear and bulk viscosities.