This is my first look at the effect that the service gap has on jet reconstruction.

To simulate the service gap in the real data, I have blanked out two barrel modules. Each module spans 0.1 units in phi and 1 unit of eta and I picked modules adjacent in phi (18 and 98 specifically). The service gap is roughly 0.1 unit in eta, so scanning across a blanked out module in phi should be approximately equivalent to scanning across the service gap in eta.

I am using the first 20 runs from Pibero's golden list. I have used the same jet finder parameters that were used in the 200 GeV 2009 data production. Since I have only blanked out modules in the barrel, I have restricted my investigation to jets in the barrel. I have placed limits of  -0.7 < detector eta < 0.7 and -0.8 < true eta < 0.8 on the jets I use in my analysis.

I have divided the analysis into five parts, the first four parts compare the average properties of jets found with all modules included and jets found with modules 18 and 98 absent. Each part contains jets which could have fired a specific trigger, L2JetHigh (dijet or random), JP1 but not L2JetHigh, Jets which didn't fire L2JetHigh or JP1 (as long as another jet in the event did fire one of those triggers), and all jets from an event in which at least one jet fired a trigger. Note: I have found an error in how I assign jets to the L2JetHigh trigger. The effect of the error was to reject jets which would have fired the AJP part of L2. The most noticeable change after the bug fix was that the phi distributions were more uniform across the jet patch boundries. I have left the old plots linked (in the strike through lines) in case people want to compare before and after the bug fix.

The plots containing jets which could have fired the L2 trigger are here.

The plots containing jets which could have fired the JP trigger are here.

The plots containing jets which didn't fire either trigger are here.

The plots containing all jets are here.

The plots containing jets which could have fired the L2 trigger are here.

The plots containing jets which could have fired the JP trigger are here.

The plots containing jets which didn't fire either trigger are here.

The plots containing all jets are here.

The above pdf files all contain 14 pages. Pages 1-3 show the Pt, eta, and phi spectra of the all module jet sample and the missing module jet sample as well as showing the difference between the two samples. Page 4 shows the average jet pt for each sample as a function of jet phi. Pages 5, 6, and 7 show the full jet Pt minus the Pt of tracks in the jet, full jet Pt minus the Pt of tower in the jet, and track pt minus tower pt respecively all as functions of jet phi. For these pages the left column is the all module jet sample and the right column is the missing module jet sample, the top row shows the full range and the bottom row shows a blown up view of the region around the missing modules. Pages 8-10 are analogous to pages 5-7 but show phi instead of pt. Pages 11-13 again have the same structure as pages 5-7 but show eta. Page 14 shows the eta-phi distributions of jets as well as towers within a jet for the all module and missing module samples. Note that in the tower eta-phi plots, only barrel towers are shown.

The last part of the analysis looks at how jet properties change on a jet by jet basis between the all module sample and the missing module sample. To do this I try to match each jet from the all module sample to a jet from the missing module sample and then look at how the jet properties differ between the matched jets. For each jet in the all modules sample I loop through all the jets in the missing modules sample and for each combination I calculate delta eta, delta phi, and R, where R is the square root of the sum of the squares of delta eta and delta phi. The missing module jet which minimizes R is the jet which matches the given all module jet. Only pairs for which both delta eta and delta phi are less than or equal to 0.7 are considered.

The plots of the jet by jet investigation are here.

The plots of the jet by jet investigation are here.

The Jet by Jet investigation pdf contains 6 pages. The first page shows the Pt, eta, phi and R differences between all pairings of all module jets and missing module jets. The second page shows the absolute value of the eta, phi, and R differences plotted on page 1. These are the quantities used to match all module jets to missing module jets. Page 3 shows the Pt, eta, phi, and R differences for each matched pair of jets. Page 4 shows the same quantities as page 3 plotted vs the phi position of the all module jet. Page 5 is the same as page 4 except that the full jet information is being used for the all module jet whereas only the tower information is being used from the missing module jet. Page 6 is the same as page 4 and 5 except that now both the all module and missing module jets are using tower information only.

Figure 1: This plot gives the phi vs eta distribution of towers in the jets. The blanked out modules are evident at phi = -0.5 . Only barrel towers are shown. This plot is from the All triggers group.

Figure 2: This plot shows the phi distribution of the jets. A number deficit can be seen in the sector with the blanked out modules.

Figure 3: This plot shows the average jet Pt as a function of jet Phi. We can see that the Pt average between -1 and 0 is lower than the the average in any other region. The enhancements in Pt at integer values of phi is most likely due to the gaps in the phi acceptance of the trigger. Jets in the gaps need more energy to fire one of the jet patches since they are sharing energy between two patches.

Figure 4: This plot shows the full jet Pt - tower only Pt vs the Phi position of the jet. The left column is jets from the all module sample and the right column is jets from the missing module sample. The top row shows the full range and the bottom row shows a blow-up of the region of interest.

Figure 5: This plot shows the full jet phi - tower only phi vs the phi position of the jet. The division of plots is the same as in figure 4.

Figure 6: This plot shows the differences in Pt, eta, phi, and R on a jet by jet basis for jets from the all modules sample which have been matched to jets from the missing module sample.