NOvA muon energy scale systematic

The systematic uncertainty on the correspondence between muon range and energy is developed for the NOvA neutrino experiment. NOvA consists of two detectors, the Near Detector at Fermilab and the Far Detector in northern Minnesota. Total errors are developed for the Near Detector, with its Muon Catcher treated separately, the Far Detector, and all combinations of correlated and uncorrelated errors between these three detectors. The absolute errors for the Near Detector (1.0%), the Far Detector (0.9%), and the fully correlated error shared by them (0.9%) are strongly dominated by Geant4's treatment of the Bethe density effect. At the Near Detector, the next biggest uncertainty is from stray hits caused by neutron capture pile-up. Other contributions are marginally significant, with the biggest, in descending order, being due to external measurements of the mean excitation energies of elements, detector mass accounting, and modification of energy loss by chemical binding. For the Muon Catcher, the absolute error is expressed as an offset instead of a percentage: 21 MeV. The density effect (at higher energies) and neutron capture pile-up (at lower energies) are the strongly dominant errors. The relative error between the Near and Far Detectors is 0.4% and is strongly dominated by neutron capture pile-up at the Near Detector, with a subdominant contribution from detector mass accounting.