- Status of leptonic x-sections (Elsa Merle) ------------------------------------------ Elsa has given a status report on the leptonics selection at various cms-energies. I) DImuons: ------- First she has shown results for 130 GeV, Lint = 6.179 pb-1 (data from 95 and 97). The x-sections are in good agreement with the SM, however, the Afb comes out high, 0.862+-0.078 (0.713 expected). There seems to be an axcess at high cos(theta*), which however is strange since the x-sections agree. At 136 GeV she has analyzed 6.376 pb-1 (95+97). There x-sections and asymmetries agree with the expectations. 161 GeV: She has added a cut in the analysis, namely a cut on the # of hits in the last HCAL pane in order to reject dielectrons which went into cracks. The resulting x-sections and asymmetries are in agreement with the expectation. 170+172 GeV: There the exclusive x-section (sqrt(s'/s)>0.85) is about 2 sigma low. 181-184 GeV : She has analyzed 57.534 pb-1, where she asked for the following RUNQ criteria: MI.AND.EF.AND.GA.AND.IT.AND.LX x-sections and asymmetry agree with what is expected. Excess in the plot Minv vs. sqrt(s'/s) : The previously reported excess (for 60 < Minv < 110GeV, sqrt(s'/s)>0.9) can be explained via missed background. Now she expects 4 BG events and 4 dimuons, and she sees 11 events in the data. At 172 GeV she observes still some excess, but this very likely is a statistical fluctuation. II) DIelectrons: ----------- A summary table of the x-sections at the various cms-energies shows globally good agreement with the SM expectations (at least for the inclusive sample, for the exclusive sample she didn't have the expected numbers). At 183 GeV (Lint=56.740 pb-1) she has used: EI.AND.EF.AND.GA.AND.IT.AND.LX.AND.LV There has been some discussion about the shown distribution of cos(theta*). The agreement between data and MC is not so good. She has found a problem in the MC, which shows an excess in the region sqrt(s'/s)>0.5 for -0.9 < cos(theta*) < 0.9. There is an excess of events with a hard photon radiated. Note that she computes s'/s w.r.t. the beam pipe. There are strong indications that there is a problem with UNIBAB. III) DItaus: ------ She has upgraded the selection, studies several decay modes of the taus independently (tau->mu antinu_mu nu_tau, tau -> hadrons + nu_tau, tau -> hadrons + nu_tau + pi0). This gives better background rejections (<9% at high energy) and higher efficiencies, up to 5% more. Work is in progress and final numbers will be shown next time. IV) Conclusions: =========== Results for Moriond should be ready. Some discussions followed about some runs at 172 GeV which are classified MAYB, but show serious problems in HCAL. These should be rejected. A mail should be sent to J.Boucrot. Ian points out that he found x-sections 5times to high when using the calorimetric selection. Elsa doesn't see that. It could be due to the fact that she applies a cut on the # of hits in the last HCAL plane. She also cuts on the # of tracks < 8 in the dimuon selection.
- Status of leptonic x-sections (Marc Swynghedauw) -------------------------------------------- For the dimuon selection he uses the package Pepidor, which is not officially released yet. s'/s is computed wrt. the beam pipe. Strange: He gets different Zfitter expectations than Elsa. To be checked. His main concern at the moment: the Bhabha Monte Carlo (UNIBAB). First he points out a possible problem of the s'-definition via the angles: s'/s can be above 0.9 even if they are not back-to-back. This occurs since the two polar angles appear as a sum (..sin(th1+th2)..), so it is enough that the sum is Pi. An additional phi requirement would be helpful. Then he showed a plot of Minv vs. Sum_of_momenta, two tracks. There one sees a big difference between data and MC, particularly the region at low Minv is empty in data, but enriched in the MC. Some checks with KORAL07 show better agreement with the data than UNIBAB. This is also a problem for the tau-selection, since Bhabhas are a background. The decay-classification is performed using the packages Idefix and Epidor. As is the case for Elsa, for the tau selection some work is still to be done. Conclusions: =========== The problem with the Bhabha MC has to be understood.
- Status of hadronic x-section (Marie-Noelle Minard) ------------------------------------------ I) Stability checks: ---------------- A previously shown plot about the dependence of the cross section on the data taking period was wrong. She has redone it and a good stability is observed. II) Selection: --------- Basic cuts are : Ntracks>4 and Minv > 50 GeV. Minv is calculated WITHOUT the photonic jet, LCAL and SICAL. (This way she kills dangerous Bhabha background from the beginning, see also next talk). The efficiency then is : 0.839 +- 0.007 +- 0.009 The second error comes from the calorimeter energy scale. She reruns the EFLOW after rescaling the calorimeter cells. The efficiency given is a reweighted one, ie., the one which turns out after reweighting the Pythia photon angle distribution. Generally better agreement with the data is observed for KORALZ compared to Pythia. The efficiency is not flat when plotted as a function of the jet angles. A cut on large jet angles is applied. The inclusive x-section is : 104.84 +- 1.56 +- 1.36 pb, expected: 104.1 pb. A control study, taking the central region only, gives an efficiency of 51.81%, and an inclusive x-section of 105.03 +- 2.19 +- 1.06. The exclusive x-section turns out to be: 25.509 +- 0.766 +- 0.410 pb, expected : 24.15 pb ===> excess is not so dramatic any more. A very efficient WW rejection can be obtained if one cuts on the thrust > 0.85. The efficiency stays almost the same, about 87 %, the WW background is reduced to 7.6%. The resulting x-section is 25.6 +- 0.73 +- 0.32 pb. She has also performed a detailed study of the backgrounds and systematics for the exclusive x-section. Dieter thinks that the second analysis should be the standard one, we should cut on thrust if this improves the measurement. He also asks for a write-up.
- Checks on hadronic x-section (Frederic Teubert) -------------------------------------- Frederic, Ian and Guenther have set up simple selection programs for dimuons and hadrons. In the hadronic case Frederic has discouvered that the excess of our high-energy x-section stems almost completely from the last bin in the thrust distribution. Scanning revealed that these are double radiative bhabha events, where both gammas convert (thus giving 6 tracks and passing the Ntrack>5 cut) (or new physics....) Again, there could be a problem with UNIBAB, Guenther will generate events with BHWIDE to check this. The problem can be avoided by cutting more strongly on the number of tracks, Ntracks>6. Then the MC thrust distribution is in agreement with data. In the end he finds a x-section of 24.71 +- 0.81 pb. Further checks will be done.
- Discussion ---------- In the discussion it is decided that we stick to a cut of sqrt(s'/s)>0.9, since this one seems to give smaller errors than 0.85, even if the other experiments use 0.85.
- Next meeting ---> 11.2.98
Guenther Dissertori
Wed January 28, 11:20:38 MET DST 1998