The expected backgrounds in our signals are evaluated by a combination of a Monte Carlo study as well as an analysis of reflection from known decays modes using data. For the normalization mode the background from is evaluated to be of events for Scenario A and events for LOWPT data. This reflection background has been evaluated by Monte Carlo, the large error is mostly due to BR's and uncertainties.
For the signal mode we evaluated the reflection background from to be events for Scenario A and events for LOWPT. Combinatorial background in mode was evaluated from data by extrapolating in the signal region under the peak the prediction of sidebands data. We get and . The consistency of this estimate was cross checked varying the cuts and using various independent background samples from different sideband definitions. The error reflects uncertainty in the background shape model (fit function) and the variation of the estimate in the different background samples.