AtlasPisaTauTauAnalysis2011

Interested in this analysis

• Zinonas Zinonos - Pisa
• Daniele Puddu - Pisa
• Vincenzo Cavasinni - Pisa
• welcome

Brother Analyses

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/MSSMHiggsToTauTauToHH2011Summer

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/MSSMHiggsToTauTauToLH2011Summer

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/MSSMHiggsToTauTauToLL2011

GRL

Information:

• Existing Good Run Lists for Data
• Good Run List Generator
• Good Run Lists Documentation, Tutorial etc

For this analysis:

• Use the data11_7TeV* xml file in the combined tauPerf GLR directory: http://atlasdqm.web.cern.ch/atlasdqm/grlgen/CombinedPerf/Tau/Tau_h_v01/

Luminosity

Information:

ATLAS Luminosity Calculator: https://atlas-datasummary.cern.ch/lumicalc/

For this analysis:

Samples

Information:

• Central Ntuple Production:
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauD3PDMaker
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauWGDPDs
• Ntuple Production tags:
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauWGDPDs#Concerning_the_different_TauD3PD
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauD3PDMaker#Interesting_Tags_of_TauD3PDMaker
• Ntuple Contents:
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauD3PDMaker#Contents_of_the_Tau_Branch_of_Co
• Tau MC Status of 7 TeV Samples:
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauMC7TeV

• Data and MC for analysis:

  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/DataMCForAnalysis
  • underatinding the production tags: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/DataMCForAnalysis#Data_and_MC_processing_campaigns
• MC Production group:
  • https://twiki.cern.ch/twiki/bin/view/AtlasProtected/AtlasProductionGroup
  • sepcific info on MC campaings: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/AtlasProductionGroup#Specific_Information_on_MC_campa

Data

Periods

Information:

https://atlas.web.cern.ch/Atlas/GROUPS/DATAPREPARATION/DQSummary/

Production

2010

2011

Tag in use: prod11082011

Monte Carlo

Monte Carlo Simulation

AMIWebSite: http://ami.in2p3.fr/opencms/opencms/AMI/www/

MC Sample	total events on AMI	Cross-section on AMI	Event Filter on AMI	PDF
mc10_7TeV.106052.PythiaZtautau.evgen.EVNT.e574	2000000	8.5400E-01 nb	1.0000E+00	MRST+fit:+arXiv:0711.2473v2+-+LOmod
mc10_7TeV.106062.McAtNloZtautau.evgen.EVNT.e598	2000000	9.5354E-01	1.0000E+00
mc10_7TeV.107054.PythiaWtaunu_incl.evgen.EVNT.e574	2000000	8.9332E+00	1.0000E+00
mc10_7TeV.108329.McAtNloWmintaunu_no_filter.evgen.EVNT.e598	1000000	3.9632E+00
mc10_7TeV.108328.McAtNloWplustaunu_no_filter.evgen.EVNT.e598	1000000	5.8403E+00
mc10_7TeV.107391.PowHegWmintaunu_incl_Pythia.evgen.EVNT.e783	1500000	3.9893E+00	1.0000E+00
mc10_7TeV.105200.T1_McAtNlo_Jimmy.evgen.EVNT.e598	15000000	1.4399E-01	5.5551E-01
mc10_7TeV.105204.TTbar_FullHad_McAtNlo_Jimmy.evgen.EVNT.e598	1200000	1.4403E-01 nb	4.4457E-01	CTEQ6.6
mc10_7TeV.105009.J0_pythia_jetjet.evgen.EVNT.e574	16400000	9.8608E+06 nb	1.0000E+00	MRST+fit:+arXiv:0711.2473v2+-+LOmod
mc10_7TeV.105010.J1_pythia_jetjet.evgen.EVNT.e574	7400000	6.7818E+05	1.0000E+00
mc10_7TeV.105011.J2_pythia_jetjet.evgen.EVNT.e574	2800000	4.0982E+04	1.0000E+00
mc10_7TeV.105012.J3_pythia_jetjet.evgen.EVNT.e574	2800000	2.1929E+03	1.0000E+00
mc10_7TeV.105013.J4_pythia_jetjet.evgen.EVNT.e574	2800000	8.7701E+01	1.0000E+00
mc10_7TeV.105014.J5_pythia_jetjet.evgen.EVNT.e574	2800000	2.3501E+00	1.0000E+00
mc10_7TeV.105015.J6_pythia_jetjet.evgen.EVNT.e574	2800000	3.3614E-02	1.0000E+00
mc10_7TeV.105016.J7_pythia_jetjet.evgen.EVNT.e574	1400000	1.3744E-04	1.0000E+00
mc10_7TeV.105017.J8_pythia_jetjet.evgen.EVNT.e574	1400000	6.2099E-09	1.0000E+00
mc10_7TeV.107670.AlpgenJimmyZtautauNp0_pt20.evgen.EVNT.e737	6620000	6.6956E-01	1.0000E+00
mc10_7TeV.107671.AlpgenJimmyZtautauNp1_pt20.evgen.EVNT.e737	1335000	1.3465E-01	1.0000E+00
mc10_7TeV.107672.AlpgenJimmyZtautauNp2_pt20.evgen.EVNT.e737	405000	4.0762E-02	1.0000E+00
mc10_7TeV.107673.AlpgenJimmyZtautauNp3_pt20.evgen.EVNT.e737	110000	1.1274E-02	1.0000E+00
mc10_7TeV.107674.AlpgenJimmyZtautauNp4_pt20.evgen.EVNT.e737	30000	2.8390E-03	1.0000E+00
mc10_7TeV.107700.AlpgenJimmyWtaunuNp0_pt20.evgen.EVNT.e600	3417000	6.9186E+00	1.0000E+00
mc10_7TeV.107702.AlpgenJimmyWtaunuNp2_pt20.evgen.EVNT.e760	3770000	3.7780E-01	1.0000E+00
mc10_7TeV.107703.AlpgenJimmyWtaunuNp3_pt20.evgen.EVNT.e760	1010000	1.0190E-01	1.0000E+00
mc10_7TeV.107704.AlpgenJimmyWtaunuNp4_pt20.evgen.EVNT.e760	250000	2.5651E-02	1.0000E+00
mc10_7TeV.107705.AlpgenJimmyWtaunuNp5_pt20.evgen.EVNT.e760	65000	6.9938E-03	1.0000E+00
mc10_7TeV.105338.HerwigVBFH120tautauhh.evgen.EVNT.e598	60000	7.5515E-05 nb	7.7044E-01	MRST+fit:+arXiv:0711.2473v2+-+LOmod
mc10_7TeV.125200.PowHegPythia_VBFH100_tautauhh.evgen.EVNT.e776	100000	1.4658E-04	1.0000E+00
mc10_7TeV.125201.PowHegPythia_VBFH105_tautauhh.evgen.EVNT.e776	100000	1.3942E-04	1.0000E+00
mc10_7TeV.125202.PowHegPythia_VBFH110_tautauhh.evgen.EVNT.e776	100000	1.3219E-04	1.0000E+00
mc10_7TeV.125203.PowHegPythia_VBFH115_tautauhh.evgen.EVNT.e776	100000	1.2613E-04	1.0000E+00
mc10_7TeV.125204.PowHegPythia_VBFH120_tautauhh.evgen.EVNT.e776	100000	1.1929E-04	1.0000E+00
mc10_7TeV.125205.PowHegPythia_VBFH125_tautauhh.evgen.EVNT.e776	100000	1.1380E-04	1.0000E+00
mc10_7TeV.125206.PowHegPythia_VBFH130_tautauhh.evgen.EVNT.e776	100000	1.0834E-04	1.0000E+00
mc10_7TeV.125207.PowHegPythia_VBFH135_tautauhh.evgen.EVNT.e776	100000	1.0336E-04	1.0000E+00
mc10_7TeV.125208.PowHegPythia_VBFH140_tautauhh.evgen.EVNT.e776	100000	9.9105E-05	1.0000E+00
mc10_7TeV.125209.PowHegPythia_VBFH145_tautauhh.evgen.EVNT.e776	100000	9.4357E-05	1.0000E+00
mc10_7TeV.125210.PowHegPythia_VBFH150_tautauhh.evgen.EVNT.e776	100000	9.0487E-05	1.0000E+00

Monte Carlo Ntuple Production

MC Sample	Location on disk	Events on Disk	Comments	Status
	/nasatlas1/AtlasData/tauPerf/superslim/prod04072011

Plot binning

Information:

• Studies: slides,
• Use a Log-type Trigger-driven binning
• pTlead, pTsublead > 40, 30 GeV: pT={20, 30, 40, 60, 90, 130, 180, 270, 390, 550, 800, 1200, 1700} GeV
• pTlead, pTsublead > 50, 40 GeV: pT={20, 30, 50, 80, 130, 200, 310, 500, 800, 1200, 2000} GeV
• pTlead, pTsublead > 40, 30 GeV: pT={20, 25, 35, 45, 60, 80, 110, 140, 180, 240, 320, 420, 550, 750, 1000, 1300, 1700} GeV

For this analysis:

• pTlead, pTsublead > 40, 30 GeV: pT={20, 30, 40, 60, 90, 130, 180, 270, 390, 550, 800, 1200, 1700} GeV

Trigger

Trigger Bank

Monte Carlo

e574_s934_s946_r2179_r2176.01-00-20

• EF_2tau16_loose
• EF_2tau20_loose
• EF_2tau29_loose
• EF_2tau29_loose1
• EF_2tau29_loose1_EFxe15_noMu
• EF_2tau29_loose1_EFxe40_noMu
• EF_2tau29_medium1
• EF_2tau38_loose

Trigger Strategy

The tau pT thresholds, for the offline analysis, obtained by the turn-on efficiency plateaus are not yet known and hence not utilized. The following hypotethical pT cuts will be used instead on a temporary basis

Data 2011:

Assume that the EF_tau29_medium1_tau20_medium1 becomes completely efficient at pT = 40 and 30 GeV for the leading and subleading tau candidate in the event

Data 2010:

Jets

Reconstruction and Calibration

Information:

JetEtMiss reccommendation: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/JetEtmissDataAnalysisRecommendationSummer2010#Recommendation_for_MET_reconstru

For physics analysis the JetEtMiss groupd recommends to use the anti-kt jet algorithm with R=0.4 or R=0.6 using topological cluster as inputs. The recommended calibration is based on the electromagnetic scale with a jet energy scale factor that depends on the transverse jet momentum and the rapidity:

Instructions how to obtain the recommended EMJES calibration from Standard Jets reconstructed are given here.

For the moment we are using the AntiKt4TopoEM jet collection as derived from the ntuple production. Soon we will move to AntiKt4LCTopoJets ( jets_* variables as from 01-00-30 )

as natural seeds to the the tau reconstruction.

For this analysis:

Use Antikt4TopoEM jets with default the kinematic variables in the D3PDs.

Taus

Tau Recommendations

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauRecommendationsWinterConf2011

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauRecommendationsSummerConf2011

Tau Reco

Information:

Tau reconstruction convention:

• tau_author=1: Calo-seeded candidate but not track-seeded
• tau_author=2: Track-seeded candidate but not calo-seeded
• tau_author=3: Seeded by both calo and track

For this analysis:

Taus are requested to be built as:

• tau_author == 1 or 3

Tau ID

Information:

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauWGOptimization2011

For this analysis:

• Prongs
  
  • tau_seedCalo_numTrack for leading: tau: 1 or 3
  • tau_seedCalo_numTrack for sub-leading: tau: 1 or 3
• tau_electronVetoMedium : must be false
• tau_muonVeto : must be false
• |tau_ charge| : 1
• id discriminant:
  • for leading tau: tau_tauLlhMedium
  • for sub-leading tau: tau_tauLlhMedium

Tau Kinematics

Principally depends on the trigger item used.

For this analysis:

• Pseudorapidity: |η|<2.5 ( assuming very energetic taus, E~p>>m(τ) )
• Transverse momentum:
  • For EF_tau29_medium1_tau20_medium1 assume the following preliminary pT cuts on the offline tau candidates
  • pT^lead > 40 GeV
  • pT^sublead > 30 GeV

Tau Lorentz Vector

Assuming for tau that its mass is negligibly smaller than its energy, E>>m, the energy four-momentum can be defined as below. Consequently, the agnular cut should be done on the tau's pseudorapidiy η and not the rapidity y.

For this analysis:

The tau for-momentum vector is defined as follows

   double eta=(*tau_eta)[j];
   double E=(*tau_Et)[j]/std::tanh(eta); 
   TLorentzVector coolTau;  coolTau.SetPtEtaPhiE( (*tau_pt)[j]/1000., eta, (*tau_phi)[j], E/1000. );

ΔR Matching and overlap removal

Information:

According to

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauD3PDMaker#Matching

the jet-to-tau and tau-to-jet matching, jet_tauAssoc_index and tau_jetAssoc_index variables correspondingly, are performed within ΔR < 0.4.

For this analysis:

A tighter jet-tau association is required with ΔR < 0.2.

Cleaning

Bad Jet Event Veto

Information:

The jet quality definitions are implemented in the functions found at

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/HowToCleanJets#Bad_jets_rel16_data

For this analysis:

Events are vetoed if a jet is found to have

• AntiKt4TopoEM jets
• jet pT > 20 GeV, |η|<4.5
• quality = MediumBad

How to do it:

• Use C++ functions from https://twiki.cern.ch/twiki/bin/view/AtlasProtected/HowToCleanJets#Bad_jets_rel16_data
• The medium bad flag for the ith-jet is then defined as follows

      isBadDefPt = pT at the EM+JES calibration scale
      bool isBadMediumUser = isBadJet( MediumBad, 
                  (*jet_AntiKt4TopoEM_emfrac)[j],
                  (*jet_AntiKt4TopoEM_hecf)[j],
                  (*jet_AntiKt4TopoEM_LArQuality)[j],
                  (*jet_AntiKt4TopoEM_HECQuality)[j],
                  (*jet_AntiKt4TopoEM_Timing)[j],
                  (*jet_AntiKt4TopoEM_sumPtTrk)[j],
                  (*jet_AntiKt4TopoEM_emscale_eta)[j],
                  isBadDefPt*1000.0,// JESPT in MeV
                  (*jet_AntiKt4TopoEM_fracSamplingMax)[j],
                  (*jet_AntiKt4TopoEM_NegativeE)[j]);

Since Chf=sumPtTrk/JESPT, both must have the same energy units (MeV).

More information:

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/HowToCleanJets

Bad Tau Jet Event Veto

Information:

Events are vetoed if a tau_jet is found to have

• tau_jet pT > 20 GeV, tau_jet |η|<2.5
• tau_author = 1 or 3
• quality = MediumBad

How to do it:

More information:

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauCleaningWinterConf2011#Jet_Cleaning_definition_for_Tau

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauCleaningWinterConf2011#rel_16_tau_object_cleaning

For this analysis:

Not used for the moment.

LAr Cleaning

LAr Errors

Check for noise bursts or data integrity errors by checking the variable larError: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/LArCleaningAndObjectQuality#LAr_Cleaning

• Throw away events if larError = 0

( 0:OK, 1:Noise Burst, 2:Data Integrity Error)

LAr Black Hole

Information:

• Starting from run 180614, missing LAr FEBs have resulted in a hole in the EM calorimeter. A report about this problem is given in this talk.
• https://atlasop.cern.ch/twiki/bin/view/Main/LArFwdDetWhiteBoard to see the list missing FEBs responsible for the LAr hole
• Jet/etmiss group is recommending a strategy to veto events if a calorimeter jets falls in the vicinity of the LAR hole. The details are explained in the jet cleaning twiki page: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/HowToCleanJets#Guidelines_for_LAr_Hole_treatment. A summary of this effect can be found at this talk.
• TauWG recommendations: https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauRecommendationsSummerConf2011#LAr_region_Veto

For this analysis:

For Data, from run 180614:

• Veto event if any preselected jet is falling in this region:
  • Jet pT>20GeV (not 15GeV)
  • -0.20 < eta < 1.65 and -0.988 < phi < -0.392
• To avoid mis-measured or mis-identified taus: when the direction of the leading track lies in the range: -0.1 < eta < 1.5 and -0.94 < phi < -0.54, the event is vetoed

For runs before 180614, the region does not need to be vetoed.

Primary vertex selection

Information:

A recommendation is to require at least one primary vertex being consistent with the beamspot position (beamspot constraint included in the data processing).

For this analysis:

Current requirement: at least one primary vertex of any type and with tracks ( is not asked any more)

Systematics

T au Systematics

Information:

https://twiki.cern.ch/twiki/bin/view/AtlasProtected/TauSystematicsSummerConf2011

For this analysis:

..

Jet systematics

MET systematics

Proposed Cutflow

1. JetTauEtMiss stream events
2. GRL
3. Trigger
4. Event cleaning
   • Primary vertex
   • Bad-jet event veto
   • Bad-tau event veto
   • LAr cleaning
     • LAr errors
     • LAr hole
5. Tau reconstruction and identification
   1. N ≥ 2 calo reconstructed taus
   2. Tau kinematic cuts |η|<2.5, pT>40,30GeV
   3. ID Taus
   4. Sharp N=2 tau candidates
   5. OS charge
6. Tagging forward-backward jets
   1. At least 2 jets in jet collection after the tau-jet overlap removal (take the 2 leading jets)
   2. Jet kinematic cuts |η|<4.4, pT>20GeV
   3. Opposite hemispheres : η(jet1) × η(jet2) < 0
   4. Eta-gap |Δη| > ...
   5. Dijet invariant mass M > ...
7. Tau centrality η(jet^-) < η(τ^-), η(τ⁺) < η( jet^-)
8. ET^miss
   1. Bisecting the tau-pair azimuthal opening angle: max[ δφ(τ⁺, ET^miss), δφ(τ^-, ET^miss) ] < δφ(τ⁺, τ^-) ≡ 0 < x1, x2< 0 in collinear approximation (ET^miss is physically connected to the tau-pair )
   2. MET > ...
9. Transverse mass M_T=[√2ET^misspT^ditau(1-cosΔφ)]^1/2
10. Balanced total transverse momentum | ∑ p_T(jet^-, τ^-, τ⁺, jet^-, ET^miss) | < ...
11. Collinear approximation cuts
    1. x1,x2 > 0.2
    2. cosθ > -0.9
12. Central Jet Veto cut (CJV)
    1. Veto event if a jet has pT > 20 GeV in |η|<3.5 and with |JVF| > 0.7
13. Mass window cut

Preliminary Comparisons

For the moment we use

1. event cleaning: at least one good VTX, avoid LAr errors, apply a MediumBadJet veto

2011	Daniele	Zinonas	Details
Total	?	?
Skimmed	885669	885669	rel11052011
GRL	776109	776109	glr
Trigger	776109	776109	EF_tau29_medium1_tau20_medium1
Event Cleaning	755416	755416	at least one good VTX, no LAr errors, MediumBadJet veto
Ditau	4433	4433	as at step 5 of cutflow
Dijet	282	282	as at steps 6.1-6.3
MET

Details

With LAr Hole Veto (pT jets > 20 GeV)

Cutflow Tables

VBF H

Z tau tau

ttbar

W tau nu

QCD jets

ggH

Data

NoTrigNoLLh->>>

NoTrigLLh->>>

QCDTopo->>>