Page 1
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 1
Discovery Potential for the Standard Model Higgs at ATLAS
La Thuile, 1-7 March, 2009
Representing the ATLAS Collaboration
Glen CowanRHUL
Les Rencontres de Physique de la Vallée d'Aoste
Page 2
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 2
Outline
IntroductionStandard Model Higgs @ LHC/ATLAS
Standard Model Higgs channels considered here: H → H → WW (*)→ eH → ZZ(*) → 4l (l = e, ) H → → ll, lh
Statistical combination of channels
Results
Conclusions and outlook
Page 3
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 3
Emphasis of this talk
Status of Higgs search recently published in: Expected Performance of the ATLAS Experiment: Detector, Trigger and Physics, arXiv:0901.0512, CERN-OPEN-2008-20.
Updated since previous studies: improved detector description andMC generators, higher order QCD and EW corrections included.
Statistical procedure for combination of channels. Systematics included via profile likelihood
In this talk, concentrate on "discovery modes": H → H → WW (*)→ e H → ZZ(*) → 4l (l = e, ) H → → ll, lh
Some modes not yet included, e.g., WW → ee, , lqq; → sensitivity will improve
Page 4
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 4
The Large Hadron ColliderCM energy 14 TeV (design)
Luminosity Low: 2 1033
(cms) High: 1034
Bunch crossing 24.95 ns
Particles/bunch 1.15 1011
Beam radius 16.7 m
Overlaid events 23 @ 1034 cms
Dipole field 8.33 Tesla
Stored energy 360 MJ/beam
Page 5
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 5
The ATLAS Detector
HCAL: Fe+scint. / Cu+LAr (10λ), σ(E)/E ≈ 50%/√E + 0.03 GeV
Muon: σ(pT)/pT ≈ 2% @ 50GeV to 10% @ 1TeV (ID+MS)
Trigger: L1 + RoI-based HLT (L2+EF)
Magnet: 2 T solenoid + toroids (0.5 T bar-rel, 1 T endcap)
Tracker: Si pixels, strips + TRT, σ(pT)/pT ≈5x10-4pT + 0.01
ECAL: Pb+LAr, σ(E)/E ≈ 10%/√E + 0.007
G. Aad et al., (The ATLAS Collaboration) J. Instrum. 3 (2008) S08003
Page 6
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 6
Higgs production at the LHC
Page 7
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 7
Vector Boson Fusion
VBF (qq → qqH) gives two forward jets,no colour connection → large rapidity gap
Additional kinematic properties give background suppressionstudied in ATLAS for: H → WW, ,
Page 8
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 8
Higgs decay modes
bb verydifficult
WW dominates for2MW < mH < 2MZ
ZZ "golden mode"
importantat low mass
Page 9
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 9
H → ZZ(*) → 4l (l = e,)
Effective in a broad mass range for mH > 130 GeV(except small gap between 2MW and 2MZ).
Lepton efficiency crucial: 4l ~ l4
Require at least one Z on shell.
Main triggers: pT(e) > 22 GeV or pT() > 20 GeV
Main background: ZZ(*) also (reducible) Zbb, tt, suppress with isolation, impact parameter, vertex constraints
Fit remaining background from sidebands of m4l.
Page 10
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 10
H → ZZ(*) → 4l sensitivity
Expected 95% CL upper limit on = / SM for 5 fbif no Higgs.
Expected discovery significance(# of ) of SM Higgs for 30 fb
Lower points include systematics
median
±1
±2
Page 11
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 11
H → with 2 jets
Exploit VBF kinematics to suppress Z+jets, tt back-grounds (Central Jet Veto).
Use → ll, lh (also study → hh).
Higgs mass reconstructed usingmissing ET and by assuming tau decay products collinear with tau.
Main triggers: pT(e) > 22 GeV or pT() > 20 GeV
Important backgrounds: Z + jets, tt
Z + jets background control: select Z → , replace by (MC)
Page 12
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 12
H → with 2 jets
Simultaneous fit of signal andQCD, Z+jets control samples: → lh gives most sensitivity:
Page 13
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 13
For 115 < mH < 140 GeV, BR ~ 0.002, but distinct signature.
Inclusive analysis based on m
Exclusive analyses (H + 0,1,2 jets) also use other variables(event topology, pT,, cos*); combined in simultaneous fit.
Main triggers: 2 photons, each pT > 20 GeV
Main backrounds: , +jet, jet+jet
57% of H→ events have atleast one photon convert (||<2.5)
Conversion reconstructionefficiency 66.4% (r < 40 cm)
vertex from ECAL pointing and tracker.
H →
Page 14
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 14
H →
Expected discovery significancefor inclusive and combination ofexclusive channels (H + 0,1,2 jets),here both fixed and floating mH.
Inclusive analysis used in currentcombination with other channels.
m resolution crucial.
Note number of entries herefor 1 fb ~ 250 → 6% stat. error per bin.
Page 15
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 15
Most important for 2MW < mH < 2MZ
Two cases: no jets (mainly gg fusion) and 2 jets (mainly VBF)
For now only WW → e (ee, lqq under study)
No mass peak because of neutrinos → use transverse mass:
H → WW(*)
where
Page 16
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 16
Simultaneous fit of:transverse mass,pT of WW,ll (angle between ll in transverse plane)
H → WW with 0 jets
Correlated W spins cause leptons to go preferentially in same direction:
Main triggers: pT(e) > 22 GeV or pT() > 20 GeV
Main background: QCD WW
ll
Page 17
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 17
H → WW with 2 jets
Main triggers: pT(e) > 22 GeV or pT() > 20 GeV
Main background: tt
Select signal region in ll, ll
Simultaneous fit of transverse mass and neural net output based on jet activity
Page 18
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 18
Statistical combination of channelsTreat systematics by means of profile likelihood method.
Considers fixed-mass hypothesis; "look-elsewhere-effect" mustbe studied separately.
Studies for individual channels also use floating-mass fits.
Approximate methods for discovery/exclusion significance.
Valid for L > 2 fb. For lower L need MC methods.
Does not represent final word on methods -- other developments ongoing (Bayesian, CLs, look-elsewhere-effect,...)
In following describe method for binned distributions; somechannels also use unbinned fits.
Page 19
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 19
Statistical model
Bin i of a given channel has ni events, expectation value is
Expected signal and background are:
is global strength parameter, common to all channels.= 0 means background only, = 1 is SM hypothesis.
btot, s, b arenuisance parameters
Page 20
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 20
The likelihood function
represents all nuisance parameters, e.g., background rate, shapes
here signal rate is only parameterof interest
The single-channel likelihood function uses Poisson modelfor events in signal and control histograms:
The full likelihood function is
There is a likelihood Li(,i) for each channel, i = 1, …, N.
data in signal histogramdata in control histogram
Page 21
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 21
Profile likelihood ratioTo test hypothesized value of , construct profile likelihood ratio:
Maximized L for given
Maximized L
Equivalently use q = 2 ln ():
data agree well with hypothesized → q small
data disagree with hypothesized → q large
Distribution of qunder assumption of related to chi-square(Wilks' theorem, approximation valid for roughly L > 2 fb):
Page 22
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 22
p-value / significance of hypothesized
Test hypothesized by givingp-value, probability to see data with ≤ compatibility with compared to data observed:
Equivalently use significance,Z, defined as equivalent numberof sigmas for a Gaussian fluctuation in one direction:
Page 23
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 23
SensitivityDiscovery:
Generate data under s+b (= 1) hypothesis;Test hypothesis = 0 → p-value → Z.
Exclusion:Generate data under background-only (= 0) hypothesis;Test hypothesis .If = 1 has p-value < 0.05 exclude mH at 95% CL.
Estimate median significance by setting data equal to expectationvalues or by using MC.
For median, can combine significances of individual channels.
For significance of e.g. a real data set, need global fit.
Page 24
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 24
Combined discovery significance
Page 25
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 25
Combined discovery significance
Approximationsin combination not always accuratefor L < 2 fb but in most cases conservative.
Discovery signficance (in colour) vs. L, mH:
Page 26
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 26
Combined 95% CL exclusion limits
For 2 fb, expected 95% CL upper limit on mH is 115 GeV.
Page 27
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 27
Combined 95% CL exclusion limits
Approximationsin combination not always accuratefor L < 2 fb but in most cases conservative.
p-value of mH (in colour) vs. L, mH:
Page 28
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 28
Summary and conclusions
Studies here only cover Standard Model Higgs channels most relevant to discovery
some modes not yet included → sensitivity will improve
Studies use NLO cross sections, improved MC generators, detailed detector simulation, improved statistical treatment.
Systematic uncertainties treated using profile likelihood.
Due to approximations used, combined results valid for L > 2 fb
For L = 2 fb: expected discovery 5 or more in 143 < mH < 179 GeV,expected upper limit on mH at 95% CL is 115 GeV
Page 29
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 29
Special thanks to
my ATLAS colleagues, especially:
Eilam Gross, Aleandro Nisati, Ketevi Assamagan, Bill Quayle, Yaquan Fang, Ricardo Goncalo, Tom LeCompte, Bruce Mellado, Marumi Kado,...
Page 30
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 30
Extra slides
Page 31
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 31
The "look-elsewhere effect"Look for Higgs at many mH values -- probability of seeing a large fluctuation for some mH increased.
Combined significance shown here relates to fixed mH.
False discovery prob enhanced by ~ mass region explored / m
For H→ and H→WW, studied by allowing mH to float in fit:
→
Page 32
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 32
Expected discovery significance
Page 33
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 33
Expected p-value of SM vs mH
Page 34
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 34
H → 4l at different mH
Page 35
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 35
H → for exclusive channelsH + 1 jet H + 2 jets
H + ET,miss + lepton H + ET,miss
Page 36
G. Cowan / RHUL Discovery Potential for the SM Higgs at ATLAS / La Thuile, 1-7 March 2009 36
Example from validation exercise: ZZ(*) → 4l
5 level
(One minus)cumulativedistributions.Band gives 68%CL limits.
½2
½2
½2
½2
Distributions of q0 for 2, 10 fb from MC compared to ½2