Determination of the Trigger Efficiency - Measurements of the Neutral Current ep Cross Sections

During data taking the decision whether to record an event for further analysis or not is done through a multi-level trigger system. Since in the highQ² NC DIS events the electron is scattered into the LAr calorimeter the most important trigger component for this analysis is the LAr trigger.

The efficiency of a trigger element TE, or a combination of trigger elements, is defined as follows:

εT E= number of events triggered by MT and TE

number of events triggered by MT (6.15) where MT is an independent monitor trigger or trigger element. The selection of monitor triggers is given in table 6.3

LAr trigger elements monitor trigger

LAr electron 1 PSNC

LAr T0 CIP T0

CIP T0 LAr T0

ToF Veto ST57 and special runs CIP Veto ST57 and special runs Table 6.3: NC trigger elements and their monitor triggers

Two mutually independent trigger efficiencies, condition A and condition B, can be combined to form the efficiency of (A OR B):

ε_AB =ε(A||B) = ε(A) + [1−ε(A)]·ε(B). (6.16)

6.8.1 LAr electron 1 TE Efficiency

The trigger level information on the calorimeter Big Towers (see section 3.11.1 for an overview of the LAr trigger system) is exploited to study the LAr electron 1 TE efficiency. The LAr electron 1 TE can be fired by both the scattered electron and the hadronic final state. The efficiency of the LAr triggering on an electron de-posit is evaluated using those neutral current events in which the hadronic final state caused the LAr electron 1 TE to fire. In turn, the efficiency for triggering on the hadronic final state is calculated with events in which an electron deposit causes the

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 100 100 100 100 100 96 100 100

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 100 100 100 100 89 100 91 87

100 100 100 100 100 100 100 100 100 100 100 100 73 100 80 58

100 100 98 100 100 100 100 99 100 100 99 100 70 100 74 34

100 100 78 100 100 100 100 100 100 100 100 100 100 100 71 99

100 100 72 100 100 100 100 100 100 99 100 100 100 100 51 100

100 100 84 100 100 100 100 100 100 100 100 100 97 100 41 99

66 100 100 99 100 100 100 100 100 100 100 100 81 100 28 100

43 100 100 100 100 100 100 100 100 100 100 100 94 100 31 100

86 100 100 100 100 100 100 100 100 100 100 100 96 100 44 73

100 100 100 100 100 100 100 100 100 100 100 100 100 100 54 25

100 100 100 100 100 100 100 100 100 100 100 100 100 100 59 31

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 79 100 100 100 76 100 100 100

100 100 100 100 100 100 100 100 97 100 100 100 88 100 100 100

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

(a)

14 14 17 17 16 16 15 15 16 17 18 15 16 17 17 18

15 17 16 19 17 17 17 17 17 19 17 16 19 19 18 16

17 17 16 17 16 17 17 17 19 20 18 20 22 19 21 19

18 22 20 20 21 19 17 20 20 22 20 22 23 23 23 20

19 20 22 24 23 21 21 22 24 22 23 25 23 25 26 21

22 24 24 24 21 23 23 22 27 24 23 25 27 26 30 24

25 25 27 28 24 25 24 25 28 29 27 29 31 31 29 30

24 30 26 30 28 28 25 27 30 32 29 29 30 30 33 29

30 31 37 33 31 31 29 29 33 35 35 34 35 34 33 35

33 33 37 37 33 34 32 31 35 38 39 37 34 39 37 34

36 37 41 39 35 40 34 37 41 45 39 43 45 43 43 40

37 38 44 44 37 42 36 42 42 45 42 46 48 49 45 47

39 44 49 46 43 44 42 41 45 50 44 51 49 49 51 48

42 47 49 51 45 47 47 52 54 49 49 52 55 52 56 53

47 47 60 57 46 46 47 46 51 56 53 54 55 57 63 53

50 51 63 55 51 53 48 47 57 60 48 60 61 58 60 53

60 59 53 62 58 63 52 61 59 59 60 59 72 64 63 66

48 62 64 65 57 60 58 57 68 70 64 64 69 60 83 65

63 62 63 64 68 55 59 66 71 69 59 68 71 67 77 64

59 62 71 68 58 68 60 61 72 71 71 74 79 74 69 89

56 66 74 66 63 80 57 61 67 66 76 71 67 63 68 77

52 67 63 72 77 71 67 63 76 65 73 70 77 70 82 70

61 68 72 74 71 62 63 68 74 67 77 80 83 75 71 75

71 75 72 75 70 55 55 69 82 66 67 82 74 77 79 81

69 73 85 83 59 71 55 64 80 66 66 79 74 61 70 74

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

(b)

Figure 6.26: The efficiency to fire the LAr electron 1 TE by the scattered electron (a) and by the hadronic final state (b), presented in az−ϕgrid, using the 2005e⁻p data.

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 100 100 100 100 100 97 100 100

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 100 100 100 100 94 100 95 92

100 100 100 100 100 100 100 100 100 100 100 100 86 100 89 78

100 100 99 100 100 100 100 100 100 100 100 100 85 100 87 66

100 100 88 100 100 100 100 100 100 100 100 100 100 100 87 100

100 100 89 100 100 100 100 100 100 100 100 100 100 100 82 100

100 100 94 100 100 100 100 100 100 100 100 100 99 100 77 100

86 100 100 100 100 100 100 100 100 100 100 100 95 100 73 100

70 100 100 100 100 100 100 100 100 100 100 100 98 100 88 100

95 100 100 100 100 100 100 100 100 100 100 100 99 100 87 91

100 100 100 100 100 100 100 100 100 100 100 100 100 100 85 92

100 100 100 100 100 100 100 100 100 100 100 100 100 100 87 84

100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100

100 100 100 100 100 100 100 100 96 100 100 100 94 100 100 100

100 100 100 100 100 100 100 100 99 100 100 100 97 100 100 100

/ deg ϕe

-150 -100 -50 0 50 100 150

/ cm LArz

-200 -150 -100 -50 0 50

Figure 6.27: The efficiency of the trigger elementLAr electron 1, presented in az−ϕ grid using the 2005e⁻pdata. The hatched areas indicate regions which are excluded because of low efficiency of the LAr electron 1 TE.

LAr electron 1 to fire.

The efficiency to fire the LAr electron 1 TE by the scattered electron is 100% ex-cept in some local regions (see figure 6.26 (a)). These regions are attributed to areas where trigger cells have been switched off due to high noise or malfunctioning hard-ware. The efficiency to fire the LAr electron 1 TE by the hadronic final state is shown in figure 6.26 (b). It is low, about 17%, in BBE (−190< z_LAr <151 cm) and rises up to ' 70% for zLAr > 0 cm. The reason for this is that the density of the energy deposit in the LAr calorimeter by the hadronic final state is higher (Lorentz boost) in the central and forward regions compared to the one in the backward re-gion. Having more energy deposited in a single Big Tower, the probability to exceed a threshold value and to fire the LAr electron 1 TE is higher.

The efficiency for triggering on hadrons is independent of the efficiency for triggering on electrons, so that the combined trigger efficiency can be estimated using equa-tion 6.16. As an example, this combined efficiency is shown in figure 6.27 for the 2005 e⁻pdata set. The efficiency is found to be very high over the bulk of the detec-tor volume. Inefficient regions (marked as hatched areas in figure 6.27) are excluded from the analysis. After these fiducial cuts, the LAr electron 1 efficiency is essen-tially 100%.

In order to account for small time dependent effects theLAr electron 1 TE efficiency is studied in a way it is discussed above and inefficient regions are excluded for each of the helicity sub-periods of the 2003-04 e⁺p and 2005 e⁻pdata taking.

6.8.2 Timing Condition

The T0 trigger elements allow the determination of the bunch crossing time. In the subtriggers the T0 requirements from the LAr calorimeter and CIP chamber are used.

The efficiency of T0 trigger elements as function of ϕe and z impact position of the scattered electron are shown for the 2003-04 e⁺p (figure 6.28) and for the 2005e⁻p (figure 6.29) data periods. The combined LAr T0 CIP T0 efficiency is close to 100%.

/ deg ϕe

-100 0 100

TEε

0.8 0.9 1

LAr T0 CIP T0

LAr T0 || CIP T0 (a)

/ cm zLAr

-200 -100 0

TEε

0.8 0.9 1

LAr T0 CIP T0

LAr T0 || CIP T0 (b)

Figure 6.28: T0 trigger efficiencies as a function of (a)ϕeand (b)zLAr for the 2003-04 e⁺pdata period.

/ deg ϕe

-100 0 100

TEε

0.8 0.9 1

LAr T0 CIP T0

LAr T0 || CIP T0 (a)

/ cm zLAr

-200 -100 0

TEε

0.8 0.9 1

LAr T0 CIP T0

LAr T0 || CIP T0 (b)

Figure 6.29: T0 trigger efficiencies as a function of (a)ϕ_eand (b)z_LAr for the 2004-05 e⁻pdata period.

6.8.3 Veto Conditions

The veto conditions in subtriggers ST67 and ST77 are described in section 3.11.1.

The veto conditions use time-of-flight (ToF) information to reject out of time back-ground events. In addition, the CIP is able to veto backback-ground from interactions in the collimators located in the beam pipe on the basis of thez vertex origin of tracks.

The signal inefficiency due to these veto conditions, i.e. the chance of rejecting good ep events, is continuously monitored with the subtrigger ST57. This monitor trigger is a copy of ST67 without the veto conditions applied. It is prescaled to keep the the rate manageable. For most of the 2003-04 e⁺p period, ST57 monitors only the CIP veto condition. For the last part of thee⁺pand the whole 2005 e⁻p part of the

running, ST57 was loosened so that it contained neither CIP nor ToF veto conditions.

Figure 6.30 shows inefficiency of the veto conditions, determined by the monitoring subtrigger ST57.

/ deg ϕ

-100 0 100

ε1 -

0 0.01 0.02

ToF Veto CIP Veto

CIP Veto & ToF Veto

(a)

/ deg ϕ

-100 0 100

ε1 -

0 0.01

0.02 ^{ToF Veto}_{CIP Veto}

CIP Veto & ToF Veto

(b)

Figure 6.30: Veto trigger efficiencies as a function of ϕe for (a)e⁺p 2003-04 and (b) for e⁻p 2005.

To estimate the inefficiency of the T oF veto conditions for the 2003-04e⁺p period, special runs without veto requirements were used:

• 367976-367979 (67.2 nb⁻¹)

• 368015-368016 (3.9 nb⁻¹)

• 368957-368990 (113.3 nb⁻¹)

Since these samples have limited statistics and the ToF rejection does not depend on the type of ep process, lowQ² NC events were used with the following selection:

• Scattered electron in SpaCal ³.

• Electron energy Eelec >14 GeV.

• Distance Rclus of the electron cluster from the z axis in r−ϕ-plane:

Rclus >20 cm.

• Q²_e >4 GeV².

• ye<0.9.

• Longitudinal momentum balance: 35 GeV < E −pz <65 GeV.

• “Central” vertex with |zvtx|<35 cm.

The obtained value for the T oFveto inefficiency, (0.46±0.06)%, for the 2003-04 e⁺p data, is then averaged with the value estimated for the last part of e⁺pperiod, using ST57.

The efficiencies of theCIP vetoandT oF_veto requirements as well as their correspond-ing errors are listed in table 6.4 for the 2003-04e⁺pand the 2005e⁻pdata sets. Each helicity sub-period is corrected for the corresponding veto inefficiency values.

3A description of the SpaCal electron finder can be found in [13].

Year Period CIP veto eff., % ToF veto eff., % Total veto eff., % 0304 e+p RH 99.79 ± 0.03 99.30± 0.25 99.09± 0.25

LH 99.56 ± 0.07 99.30± 0.25 98.86± 0.26 2005 e−p RH 99.44 ± 0.07 99.09± 0.09 98.54± 0.11 LH 99.49 ± 0.04 99.44± 0.04 98.93± 0.06

Table 6.4: Veto efficiencies.

6.8.4 Trigger Efficiency

TheLAr electron 1 TE and timing conditions are the core of the high Q² NC trigger used in this analysis. It is required that the trigger is fully efficient for the selected data. The regions in which this is not the case are excluded as indicated on the figure 6.31.

The efficiency of the veto conditions, which are introduced in the trigger to reject non-ep background, does not depend on the epphysics process. The small inefficien-cies observed due to these requirements are corrected for.

An uncorrelated systematic uncertainty of 0.5% is attributed to the trigger efficiency.

Im Dokument Measurements of the Neutral Current ep Cross Sections Using Longitudinally Polarised Lepton Beams at HERAII (Seite 115-120)