with ATLAS A I Etienvre for the ATLAS Collaboration Introduction Motivations for a precise top quark mass measurement EW precision observables depend on the value of the top mass ID: 1038356
Download Presentation The PPT/PDF document "Top mass measurement" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
1. Top mass measurement with ATLASA.-I. Etienvre, for the ATLAS Collaboration
2. IntroductionMotivations for a precise top quark mass measurement:EW precision observables depend on the value of the top mass a precise top mass measurement is needed to: Perform consistency tests of the Standard Modelconstrain the Higgs mass within the Standard ModelSearch for new physics beyond the Standard Model July 2010
3. Top mass measurement with ATLASMeasurement :performed in the l(e,µ)+jets channel Best compromise between BR and S/BBased on template methods:Definition of an estimator sensitive to mtop Produce (MC) templates of it for various mtopFit templates for signal and background 2 PDFsLikelihood fit to the data top mass measurement Results shown today:From 35 pb-1 recorded by ATLAS at s = 7 TeVThree complementary template methods reference: ATL-CONF-2011-033 Alternative measurement: top mass extraction from ttbar cross section measurement : reference: ATL-CONF-2011-054
4. Event selectionLepton + jets standard event selection:Single lepton (e/µ) trigger firedExactly one isolated lepton:within good detector acceptancepT > 20 GeV Transverse mass (ln)/ Etmiss (QCD rejection)e channel: MT (ln) > 25 GeVETmiss > 35 GeVµ channel: MT (ln) + ETmiss > 60 GeV ETmiss > 20 GeVJets:≥ 4 jets, pT > 25 GeV, |h| < 2.5≥ 1 b-tagged jetAfter all cuts, S/B is of the order of 5 Dominant background: QCD, W+ HF
5. R32 template method Principle: Observable chosen : Reduces sensitivity to Jet Energy Scale uncertaintyHadronic top mass reconstruction:Jet triplet maximizing pT:Additional cuts: W window : 60 < mW < 100 GeVVeto 2b jets in tripletR32 template parameterizationFor signal (includes single top) : For background Does not depend on mtop e channelµ channelµ channel
6. Event yieldsQCD estimate data drivenGood agreement between data and Monte Carlo
7. R32 template methodTop mass measurementLikelihood fit performed on 35 pb-1Stat. uncertainties only in these plotsSystematic uncertainties: see next slidestat.syst.e channelµ channel
8. Systematic uncertaintiesSystematic uncertainties dominated by ISR/FSR, b-JES and JESStat. uncertainy predicted by pseudo-exp, consistent with observation
9. 2D template methodSimultaneous fit of mtop and global Jet Scale Factor (JSF):2 observables are chosen: mjj = mWreco = f(JSF)Mtopreco =f(JSF, mtop )Final state reconstruction slightly different:Light jet pair (W) chosen by minimizing:Light jet energies not rescaled for the W reconstructionBut rescaled for the top mass reconstructionHadronic top reconstruction : jet triplet maximizing pTµ channel
10. 2D template methodMeasurement with 35 pb-1 fitted Jet Scale Factor : e channel = , µ channel: stat.syst.µ channel
11. 1D template method with kinematic fitKinematic likelihood fitInformation from the entire event used to reconstruct both sides of the ttbar decay with a kinematic likelihood fit:TF relate reconstructed objects energy and direction to those of their parent partons:Lower stat. uncertainty but no constrain on JES Measurement:stat.syst.
12. Mass extraction from cross sectionExploits the dependency of the top pair cross section wrt top mass (assuming mtopMC = mtoppole ):Using most accurate ttbar cross section single measurementBased on 3 different theoretical calculations 13% uncertainty on measured cross section 6 GeV uncertainty on mtop theory uncertainty (scales + PDF) 4 GeV uncertainy on mtop
13. Conclusion3 template methods have been used for a direct top mass measurement:Consistent results Complementary Indirect measurement (extraction from ttbar cross section): consistent with direct measurementsWith 1 fb-1 : stat. uncertainty ≤ 1 GeV Systematic become crucial
14. OutlookFirst distributions of mjjb in the l+jets channel with 0.7 fb-1
15. BACK UP
16. ISR/FSR systematic uncertainty estimateATLAS has adopted a very conservative approach:6 samples have been generated (AcerMC + Herwig)enhanced ISR, or FSR, or both reduced ISR, or FSR, or bothParameters varied:ISR: PARP(67) and PARP(64)FSR: PARP(72) and PARJ(82)Likelihood fit performed:For each datasetMaximum difference = systematic uncertaintyISR/FSR studies ongoingMeasurement in data foreseen