Preparations for SUSY searches at the LHC Alex Tapper UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Outline Search strategy Examples with 70 nb Hadronic searches Leptonic searches (single and di-lepton) Photon searches (di-photon) Summary and outlook 2 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Introduction/bibliography Documents for ICHEP on preparations for SUSY searches at LHC ATLAS Collab., Early supersymmetry searches in channels with jets and missing transverse momentum with the ATLAS Detector (ATLAS-CONF-20-065) ATLAS Collab., Early supersymmetry searches with jets, missing transverse momentum and one or more leptons with the ATLAS Detector (ATLAS- CONF-20-066) CMS Collab., Performance of Methods for Data-Driven Background Estimation in SUSY Searches (CMS-SUS--001) Not planning to show Early supersymmetry searches in events with missing transverse energy and b- jets with the ATLAS detector (ATLAS-CONF-20-079) Prospects for Supersymmetry discovery based on inclusive searches at a 7 TeV centre-of-mass energy with the ATLAS detector (ATL-PHYS-PUB-20-0) The CMS physics reach in searches at 7 TeV (CMS-NOTE-20-008) 3 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search strategy (what and how?) Production Squark and gluino expected to dominate Strong production so high cross section Cross section depends only on masses Approx. independent of SUSY model 4 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search strategy (what and how?) Production Squark and gluino expected to dominate Strong production so high cross section Cross section depends only on masses Approx. independent of SUSY model Decay Details of decay chain depend on SUSY model (mass spectra, branching ratios, etc.) Assume R P conserved decay to lightest SUSY particle (LSP) Assume squarks and gluinos are heavy long decay chains Signatures MET from LSPs, high-e T jets and leptons from long decay chain Focus on robust and simple signatures Common to wide variety of models Let Standard Model background and detector performance define searches not models 5 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search strategy (MC example) How might such a generic search look? ATL-PHYS-PUB-2009-084 Simple selection categorise events by numbers of leptons and jets Jet ET > 0 (40) GeV ΔΦ(jeti,MET) > 0.2 rad Lepton ET > 20 () GeV MET > 80 GeV Meff = ΣET jet + ΣET lep + MET MET > 0.2-0.3 x Meff ST>0.2 MT > 0 GeV Good S/B for most channels (200 pb @ TeV centre-of-mass) but... Backgrounds straight from Monte Carlo Measuring backgrounds is the key 6 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Backgrounds Physics Standard Model processes that give the same signatures as SUSY Cannot rely on Monte Carlo predictions measure in data Detector effects Detector noise, mis-measurements etc. that generate MET or extra jets Commissioning and calibration results shown by Jon and Jim Beam related Beam-halo muons (and cosmic-ray muons), beam-gas events Data and simulation already measure in situ too 7 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (MC example) ATL-PHYS-PUB-2009-084 All-hadronic search highly sensitive to SUSY, but suffers from many backgrounds 8 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (step I) Simple (ignoring trigger, cleaning.) jet cuts (anti-kt R=0.4) Leading jet ET>70 GeV Other jets ET>30 GeV Veto isolated leptons (PT> GeV) QCD MC normalised to data in two jet channel (uncertainty neglected) 9 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (QCD) Mis-measurement of a jet leads to MET along the jet axis Remove with ΔΦ(jeti,MET) > 0.2 rad arxiv:0901.0512 (2009) UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (QCD) Mis-measurement of a jet leads to MET along the jet axis Remove with ΔΦ(jeti,MET) > 0.2 rad 11 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (final) Further selection MET>40 GeV MET/M eff>0.3(0.2) 12 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (αt) jet LSP LSP PRL1:221803 (2008) & CMS-PAS-SUS-09-001 jet jet jet α T = E T j 2 M T j1 j 2 = E T j 2 / E T j1 2(1 cosδϕ) A novel approach combining angular and energy measurements No dependence on MET robust for early LHC running Originally proposed for di-jet events generalised up to 6 jets Perfectly balanced events have αt=0.5 (cut at αt>0.55) Mis-measurement of either jet leads to lower values 13 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (αt) jet jet jet LSP LSP jet Events / 0.05 3 2 PRL1:221803 (2008) & CMS-PAS-SUS-09-001 CMS preliminary s = 7 TeV L = 11 nb 2 PF Jets HT>120 GeV Data Pythia8 QCD Events / 0.05 4 3 2 CMS preliminary s = 7 TeV L = 11 nb >=3 PF Jets HT>120 GeV Data Pythia8 QCD α T = E T j 2 M T j1 j 2 = E T j 2 / E T j1 2(1 cosδϕ) data/sim 1 0 0.5 1 1.5 2 2.5 3 α T 2 1 0 0 0.5 1 1.5 2 2.5 3 α T A novel approach combining angular and energy measurements No dependence on MET robust for early LHC running Originally proposed for di-jet events generalised up to 6 jets Perfectly balanced events have αt=0.5 (cut at αt>0.55) Mis-measurement of either jet leads to lower values data/sim 1 0 0.5 1 1.5 2 2.5 3 α T 2 1 0 0 0.5 1 1.5 2 2.5 3 α T 14 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (centrality) Heavy particles preferentially produced centrally Use forward regions as background control region dn d! N 1 " 0.45 0.4 0.35 0.3 0.25 CMS preliminary LM0 LM1 tt W Z QCD MadGraph 0.2 0.15 0.1 0.05 0 0 0.5 1 1.5 2 2.5 3! leading jet 15 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (centrality) Heavy particles preferentially produced centrally Use forward regions as background control region dn d! N 1 " 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 CMS preliminary LM0 LM1 tt W 0 0 0.5 1 1.5 2 2.5 3! leading jet Z QCD MadGraph > 0.55) < 0.55) # events(" T # events (" T -3 # H T > 450 GeV/c 0.2 H 0.18 T > 350 GeV/c 0.16 300 < H T < 350 GeV/c 0.14 0.12 CMS preliminary 0.1 0.08 0.06 0.04 0.02 0 0 1 2 3! leading jet 16 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (centrality) Heavy particles preferentially produced centrally Use forward regions as background control region dn d! N 1 " 0.45 0.4 0.35 0.3 0.25 0.2 0.15 0.1 0.05 CMS preliminary LM0 LM1 tt W 0 0 0.5 1 1.5 2 2.5 3! leading jet Z QCD MadGraph > 0.55) < 0.55) # events(" T # events (" T -3 # H T > 450 GeV/c 0.2 H 0.18 T > 350 GeV/c 300 < H 0.16 T < 350 GeV/c 0.14 0.12 CMS preliminary 0.1 0.08 0.06 0.04 0.02 0 0 1 2 3! leading jet 17 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
All-hadronic search (centrality) Heavy particles preferentially produced centrally Use forward regions as background control region dn d! N 1 " 0.45 0.4 0.35 0.3 0.25 0.2 0.15 CMS preliminary LM0 LM1 tt W Z QCD MadGraph > 0.55) f(" T 0.008 0.007 0.006 0.005 0.004 0.003 0.002 H T H T > 160 GeV > 160 GeV: removed jets CMS preliminary s = 7 TeV, 54 nb 0.1 0.05 0.001 0 0 0.5 1 1.5 2 2.5 3! leading jet 0 0 0.5 1 1.5 2 2.5 3! leading jet 18 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search (and discovery) variables Barr and Gwenlan PRD80:074007,2009. 19 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Other search variables 20 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
QCD control variables Events (normalized) / 0.1 0.25 0.2 0.15 0.1 0.05 CMS Preliminary s = 7 TeV 15 nb EmulatedMHT > 50 GeV 30 GeV < EmulatedMHT < 50 GeV 20 GeV < EmulatedMHT < 30 GeV Multijet data, MHT > 30 GeV Multijet MC, MHT > 30GeV 0 0 1 2 3 Δφ(MPT,MHT) Play independent detectors off against each other Normalized Entries / 0.13 1 1 2 3 160 > H T GeV > 120 H T GeV > 160 160 > H T GeV > 120: jet removed H T GeV > 160: jet removed 2 jets s = 7 TeV, 12 nb CMS preliminary 0 0.5 1 1.5 2 2.5 3 * 1 21 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Single-lepton search (MC example) ATL-PHYS-PUB-2009-084 Requiring one lepton (e or µ) suppresses QCD background powerfully Highly sensitive to SUSY Backgrounds come from Standard Model processes with neutrinos real MET In particular top and W decays 22 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Single-lepton search (part I) Simple cuts (once again too lazy to list cleaning, triggers...) One isolated lepton with P T>20 GeV At least two jets E T>30 GeV QCD MC normalised to data at MET < 40 GeV and MT < 40 GeV Uncertainty 50% from fake rate study comparison with data 23 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Single-lepton search (W bkgd) Require MET>30 GeV remove QCD background Dominated by W+jets (no ttbar yet in this dataset) Normalise W+jets in 40 < MT < 80 GeV and 30 < MET < 50 GeV 50% uncertainty assumed and propagated to high MT and MET 60% uncertainty assume for Z+jets 24 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Single-lepton search (final) Further selection MET>30 GeV M T>0 GeV 25 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Single-lepton searches (future) So far not categorised by number of jets, for the future 26 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Modelling MET resolution Aim to measure fake MET tail for single-lepton + jets + MET search Measure MET templates in multi-jet QCD events and categorise them in Njets and HT For each γ/z event look for the corresponding template Separate procedure for real MET CMS Preliminary, 65 nb, s = 7 TeV 0 < H T < 300 GeV 4 N J 300 < H T < 350 4 N J 350 < H T < 400 4 N J 400 < H T < 500 4 N J 500 < H T < 00 4 N J -2-2 -2-2 -2-3 0 20 40 60 80 MET (GeV) 0 < H T < 200 GeV = 3 N J -3 0 20 40 60 80 MET (GeV) 200 < H T < 250 = 3 N J -3 0 20 40 60 80 MET (GeV) 250 < H T < 300 = 3 N J -3 0 20 40 60 80 MET (GeV) 300 < H T < 400 = 3 N J -3 0 20 40 60 80 MET (GeV) 400 < H T < 00 = 3 N J -2-2 -2-2 -2-3 0 20 40 60 80 MET (GeV) -3 0 20 40 60 80 MET (GeV) -3 0 20 40 60 80 MET (GeV) -3 0 20 40 60 80 MET (GeV) -3 0 20 40 60 80 MET (GeV) 27 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Modelling MET resolution Aim to measure fake MET tail for single-lepton + jets + MET search Measure MET templates in multi-jet QCD events and categorise them in Njets and HT For each γ/z event look for the corresponding template Separate procedure for real MET Events / 5 GeV 1 CMS preliminary, 65 nb " +! at 7 TeV 3 jets data Template prediction 0 20 30 40 50 60 70 MET (GeV) 28 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Determining QCD background QCD small background (even smaller for µ) but with large uncertainty Example of data-driven template fit to relative isolation distribution Good closure in data method works 29 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Di-lepton searches (MC example) ATL-PHYS-PUB-2009-084 Low yields but very interesting properties Same sign searches Very low Standard Model background rate Backgrounds from charge mis-identified top events (QCD in τ channel) Opposite sign Use opposite-sign, opposite-flavour sample to subtract SM background 30 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Di-lepton searches First look at the MET distributions for di-leptons At least two muons PT1>20 GeV PT2> GeV Mll>5 GeV Normalise QCD MC to data in 5 < Mll < 15 GeV and MET < 15 GeV 0% uncertainty assumed on W and QCD backgrounds and 60% for Z Good description by Monte Carlo (so far as one can tell ) 31 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search with di-photon events -4$+5* 6 3.?(7!JB$;)9)+8BK (?!!!!789:;$!<)5=!#(3> / 3. 3. $!!@8>A,BCD+E!-*5)985$ 'C58;!@8>A,BCD+E!-*5)985$ FCB98;)G85)C+!H+>$B5I 3!3 3. Di-photons + MET search 3.!/. /. 0. 1. 2. 3.. 3/. 30. 31. 32. /.. ()**)+,!-!"#$%& ' Background is dominated by mis-measured QCD fake MET MET resolution determined by hadronic recoil Predict using control samples (now fake-fake later Z e + e - ) 32 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Search with di-photon events Events 50 45 40 35 Norm. Region CMS Preliminary!! Candidate Sample s=7 TeV, 52.1 nb Total Uncert., QCD background 30 Correlated Uncert. 25 20 15 5 Di-photons + MET search Background is dominated by mis-measured QCD fake MET MET resolution determined by hadronic recoil Predict using control samples (now fake-fake later Z e + e - ) 4 observed (MET>20 GeV) 4.2 ± 1.5 events predicted. 0 0 5 15 20 25 30 35 40 45 50 Missing E (GeV) T 33 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.
Summary and outlook Early searches based on robust generic signatures Sensitive as possible to a variety of new physics models Detectors in great shape already and ready for searches Electrons, muons, jets, MET, taus and b-tagging all available to us A wide range of data-driven techniques developed to measure efficiencies and backgrounds Will have something to say with 20 data in many channels 1fb 2011 dataset will give us huge discovery potential! 34 UK HEP Forum: LHC First Results and Outlook, September 20-21 20.