Towards a Systematic Uncertainty Model for the Powheg bb4 Monte Carlo sample

This note presents a study of the ATLAS Monte Carlo (MC) sample describing pp → b b̄ ⁺ ⁻ ν_ ν̄_ events at next-to-leading order in quantum chromodynamics. Events are generated using the Powheg Box Res implementation of the bb4 process and are interfaced with Pythia 8 for parton showering and hadronization. The nominal ATLAS bb4 MC sample is compared to predictions from Powheg+Pythia 8 for top-quark pair production (tt̄) and single-top production in association with a W-boson ($tW$). The bb4 MC sample is intended to reduce modelling uncertainties in precision top-quark physics analyses with respect to the modelling uncertainties associated with the tt̄+tW MC predictions. The impact of theoretical modeling variations is evaluated and compared to the corresponding uncertainty estimate in the tt̄ simulation. The studied uncertainties include matrix-element scale variations, changes in the Powheg+Pythia 8 matching parameters, modification of the Pythia 8 final-state splitting kernel via μR variations, the use of parton distribution function (PDF) sets with an alternative value of αS(MZ) and the inclusion of inverse width corrections in bb4. These studies support the development of a systematic uncertainty model for the bb4 simulation.