Run-13 Luminosity Effect on W Efficiency

 To Study the luminosity effect on the w efficiencies for Run 13, I first plotted the W efficiency (W+) in this case as a function of the zdc rate, which was obtained through  zdcRate=wMK->mMuDstMaker->muDst()->event()->runInfo().zdcCoincidenceRate(), for Run-13 periods 1 and 2. Both of these periods agree very well with each other, as can be seen in figure Run13-ZDCEff.pdf. 

I think what is leading to the drop in the efficiency between periods 1 and 2 is just the amount of high luminosity that was collected. If you look at figure Run13-ZDCRate.pdf, I have plotted the zdc distribution for periods 1 (red histogram) and 2 (blue histogram). Period 2 sits higher (~340 kHz) in the zdc distribution than period 1 does (~297 kHz). In addition to this if we calculate what percentage of events were collected at higher zdc rates (>300 kHz), we see that nearly 80% of period 2 data was taken above a zdc rate of 300 kHz, while only about 50% of events collected during period 1 are larger than 300 kHz. 
I think the fact that period 2 had a larger zdc rate and collected much more higher luminosity data than period 1 leads to this decrease in the W efficiencies when going from period 1 to period 2.