Direct measurement of backgrounds using reactor-off data in Double Chooz

Double Chooz is unique among modern reactor-based neutrino experiments studying (nu) over bar (e) disappearance in that data can be collected with all reactors off. In this paper, we present data from 7.53 days of reactor-off running. Applying the same selection criteria as used in the Double Chooz reactor-on oscillation analysis, a measured background rate of 1.0 +/- 0.4 events/day is obtained. The background model for accidentals, cosmogenic beta-n-emitting isotopes, fast neutrons from cosmic muons, and stopped-mu decays used in the oscillation analysis is demonstrated to be correct within the uncertainties. Kinematic distributions of the events, which are dominantly cosmic-ray-produced correlated-background events, are provided. The background rates are scaled to the shielding depths of two other reactor-based oscillation experiments, Daya Bay and RENO.