1. First of all, you may normalize/scale the spectrum to compare adcsum spectra. You may get more or less entries due to many factors, like different prescale factors and cross sections.
2. It is better to look at the pedestal unsuppresed spectrum to look at the whole distribution. The current spectrum might give us misleading information about the pedestal.
3. What to expect for different polarities: We are mostly detecting protons and pions for positive polarity. And we are mostly detecting pions and electrons for negative polarity. Both pions and electrons will fire the aergoel cherenkov detector, and electrons tend to produce a slightly higher adcsum value. The protons are not supposed to fire the aerogel cherenkov. Their adcsum values are mostly in the pedestal and the single photon-electron peak. Therefore, if we are changing polarity from negative to positive, we will expect to see relative higher pedestal and single photon-electron peak.
4. Following the discussion above, I would guess the black curve in your figure 2 is for negative polarity while the shaded reference spectrum is for positive polarity.
5. If my above guess is wrong, I need to think about it again.
A copy of this log entry has been emailed to: ycwang@jlab.org,bogdanw@jlab.org