This page details my run by run QA of the 2009 200GeV jet trees. Only production2009_200Gev_Single runs in the first priority list are considered here. This page superceedes the Jet Tree QA post from May 21, 2010.

Low Statistics Sample:

The plots in this section are all generated from the first pass jet trees. For these trees, many runs exhibit a large deficit of events as compared to what we would expect to see from the run log pages. A brief discription of this problem along with its probable cause and solution can be found here. I am proceeding on the assumption that the lost events are randomly distributed and do not introduce any systematic bias into the quantities I will look at. Of course I will repeat these studies when the jet trees are rerun and compare to see if the above assumption holds.

All QA plots from the 628 first priority production2009_200Gev_Single runs can be found here. (Note that some jobs didn't finish so there may be fewer than 628 runs). The list mapping run index to actual run number can be found here.

I have also split the QA plots into Reversed Full Field and Full Field sections for easier viewing. The RFF plots can be found here with its mapping file and the FF plots can be found here with its mapping file.

Below I show selected plots of interest as well as plots which may show problems:

Trigger Ratios:

Figure 1: The top panel of this plot shows the ratio of the number of L2JetHigh triggers obtained from the jet trees vs the number of BBCMB_Cat2 triggers (L2JetHigh's luminosity counter) obtained from the jet trees. The bottom panel shows the ratio of the number of L2JetHigh and BBCMB_Cat2 triggers obtained from the jet trees vs the number of those triggers we would expect from the run log page.

Figure 2: This plot conains the same information as fig 1 but for the JP1 trigger instead of the L2JetHigh trigger.

There are 5 runs which deviate a significant amount from the remainder of the runs. These five are run index 184-187 and 277. These indices correspond to runs 10141008, 10141010, 10141011, 10141013, and 10144038.

Vertex Position:

Figure 3: This plot shows the average position of the highest ranked vertex for each run.

The plots of the x and y vertices show obvious discontinuities, these correspond to the switch from RFF to FF. In the x plot, the last point to the left of the discontinuity at about 0.35 is run index 336 which corresponds to run 10146091. This is the last RFF production2009_200Gev_Single run in the first priority run list. The first point to the right of the discontinuity at about 0.25 is run index 337 which corresponds to run 10148002. This is the first FF production2009_200Gev_Single run in the first priority run list. In the y plot, the last point to the left of the discontinuity at about -0.06 is again run index 336, the points then jump to about -0.04 starting with run index 337 and continuing to 347 which corresponds to run 10148035, finally, the points jump to about 0.03 starting with run index 349 which corresponds to run 10149012.

The plot of z vertex position shows several groups of runs with anomously low values; I have not investigated these yet.

Fired Jet Patches:

Figure 4: This plot shows the average number of Barrel, Endcap, and Overlap jet patches which fire per event. An enhancement in the number of Endcap JP fired can be seen starting at run index 62 which corresponds to run 10134024 and running to about run index 120 which corresponds to run 10136080. It should also be noted that the last run before the enhancement is 10129011, so there is a ~5 day gap.

This enhancement can also be seen in the plot of the average number of endcap towers in a jet found here. It is also visible in the plot showing the average jet pseudorapidity found here. Finally, it is somewhat visible in the plot of average track pseudorapidity found here.