Download a postscript version of this plot here.
In this plot, the x axis is the energy that appears in the BCAL EM section, binned up in 10 GeV bins. This energy includes both multiple brem and pair energy as predicted by the beam monte carlo and physics energy predicted by Lucifer and Twister.
The y axis is the mean of the difference between the apparent photon energy (E_app = E_silicon - E_resh) and the actual energy of the lead photon E_lead, average taken for each bin of BCAL EM energy. The errors in points on this axis are the sigma of the mean for the distribution in each bin.
Characteristics of a linear fit to all but the first few points are given. This fit suggests very strongly that we can make a correction to what we think the photon energy is using the BCAL EM energy.
However, the scatter plot from which these data were taken shows that there is a small population of events with significant amounts of BCAL EM energy and a small E_app-E_lead. This is unsurprising, of course, since spillover or a few far forward pi-zeros can deposit EM energy in the BCAL. Correcting the photon energy in those events for BCAL EM energy would be a mistake.
BCAL hadronic energy does not correlate with E_app - E_lead. We are checking to see if the population of events with large BCAL EM and E_app-E_lead near zero also has large BCAL hadronic energy, which would provide a check on whether or not to perform the correction.
At this point, we do not suggest an exact form for the correction. We anticipate recommending an explicit correction within a few weeks.
Greg Morrow morrow@physics.rice.edu