1: Electron identification efficiencies are measured using a combination
2: of data and a {\sc geant}-based \cite{geant} simulation of the D0 detector.
3: The electron identification efficiencies
4: are measured from \zgamma~data.
5: %are measured from \zgamma~events
6: %and are parameterized in terms of the electron
7: %transverse energy and, for some variables,
8: %the vertex position along the beam axis or electron incident angle.
9: The dependence of the overall selection efficiency on the \zgamma~boson $q_T$ is parameterized
10: from the {\sc geant} simulation. A measurement of this shape
11: from the data agrees well with the simulation within statistical uncertainties.
12:
13: The dominant backgrounds are from photon plus jet events and
14: di-jet events, with photons and jets misidentified as electrons. The kinematic
15: properties of these events are obtained from events that satisfy most of the \zgamma~selection
16: criteria, but fail the electron shower shape requirement.
17: The normalization of the background is obtained by fitting
18: to a sum of a signal shape obtained from a parameterized simulation
19: of the detector response and the invariant mass distribution from the background sample
20: to the invariant mass distribution of the data sample. The background fractions are
21: (\bckcccc$\pm$\bckccccerr)\%,
22: (\bckccec$\pm$\bckccecerr)\%, and
23: (\bckecec$\pm$\bckececerr)\%~
24: for CC, CE, and EE events respectively. Other backgrounds are negligible.
25:
26: The data are corrected for acceptances within a range of generated \zgamma~masses of 40 to 200 GeV/$c^{2}$,
27: and for selection efficiencies using a parameterized simulation. We
28: use {\sc ResBos} \cite{Yuan} as the event generator which incorporates the resummation calculation
29: in $b$-space using the BLNY parameterization for low $q_T$ and a NLO
30: pQCD calculation for high $q_T$. We use {\sc photos} \cite{photos} to simulate the effects of
31: final state photon radiation.
32: The overall acceptance times efficiency falls slowly from a value of
33: $\acceptptzero$ at low $q_T$ to a minimum of $\acceptmin$
34: at $q_T=$ \ptacceptmin ~GeV$/c$ and slowly increases for larger $q_{T}$.
35:
36: The measured spectrum is further corrected for detector resolution effects using
37: the {\sc run} (Regularized Unfolding) program \cite{RUN} to obtain the true differential cross
38: section. Its performance was verified by comparing the true and unfolded
39: spectrum generated using pseudo-experiments.
40: The measured $Z$ $q_T$ resolution is about 2 GeV/$c$; the bin width we choose
41: is 2.5 GeV/$c$ for $q_T<30$ GeV/$c$. The typical correlation
42: between adjacent bins is around 30\%. Due to limited statistics, the chosen bin width
43: is 10 GeV/$c$ for $30<q_T<100$ GeV/$c$ and 40 GeV/$c$ for $100<q_T<260$ GeV/$c$.
44:
45: Systematic uncertainties on the unfolded $q_{T}$ spectrum
46: arise from uncertainties on the electron energy calibration,
47: the electron energy resolution, the dependence of the overall selection efficiency
48: on $q_T$, and the effect of parton distribution functions (PDFs) on the acceptance.
49: The uncertainties on the unfolded spectrum are estimated
50: from the resulting change when the smearing parameters are varied within their uncertainties.
51: CTEQ 6.1M is used as the default PDF. Uncertainties due to the PDFs
52: are estimated using the procedure described in Ref.~\cite{cteq}. The uncertainty due
53: to the choice of unfolding parameters in the {\sc run} program
54: is also estimated and included in the final systematic uncertainty.
55:
56:
57: