BEMC status was queried every hour during the extent of the st_gamma data production stream (10125059 - 10181005) using the StBemcTablesWriter (in StEmcUtil/database).  The resulting histogram was used to select intervals where the BEMC status was relatively constant.  The individual intervals were then weighted by computing the number of st_gamma stream events falling into the interval.

The resulting intervals are

Update: The TPC status tables are bad during Interval IV.  Removing IV, renaming V to IV, and readjusting the weights gives

Update: Surprise!  The TpcRS runs into overflow problems during Interval II that prevent the proper use of the subsequent MuDsts.  As it's no larger than the previously removed interval, it is simply dropped.  Renaming and readjusting (also using a more up to date QA'd run list for the luminosity weighting) gives