After a comment from Carl [and subsequent clarification from Pibero] it became apparent the "Number of Events" plot in my previous post was nonsensical. A scan of the μDST's as well as the trees revealed the problem was in my coding logic. Here, I post updated plots having changed the logic; and the result is more sensical.

As before, I use the following series of cuts:

• vertex rank > 0
• |z_vertex| < 90 cm
• R_T < 0.98

Also, for the various trigger selections, I require that the trigger did fire. I do not, yet, require a matching to a valid trigger patch. That is necessary and will follow, shortly.

Figure 1

What is plotted, here, now, is the number of events satisfying the vertex cuts and containing a jet [prior to the R_T cut]. This seems to make far more sense than before. Note that in addition to losing events from the cuts, at least two trees show some errors and should be rerun in the future.

Figure 2: JP0||JP1||JP2*L2JetHigh||AJP

Here, we see the updated figures for JP0, JP1, JP2*L2JetHigh, and AJP. Slight differences from the previous updates should be due to rerunning of the trees with at least two corrupted files.

Figure 3: BHT1||BHT2||BHT0*VPDMB

Again, we see the updated figures for BHT1, BHT2, and BHT0*VPDMB.

Figure 4: BBCMB||VPDMB

Finally, we see the updated figures for BBCMB and VPDMB.

Next Steps

Renee has suggested prior to a full pass on the μDST's, I should do some QA of all runs, e.g. Trigger rate vs. Run and N_{tow above threshold}/event vs. Run. The thought is that this may aid in locating runs with hot calorimeter towers.