Post on 08-Jul-2020
Status of Standard Model Higgs Searches ATLAS
Scott SnyderFor the ATLAS Collaboration
Brookhaven National Laboratory, Upton, NY, USA
Dec 20, 2011BNL
IntroductionSM Higgs search channels
I WW → `ν`νI γγI ZZ → 4`
Combination
Conclusions/outlook
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 1 / 46
LHC
LHC performance has greatlyexceeded expectations!
2011 goal set in spring
Goal for 2011: 1 fb−1
Accomplished: ∼ 5 fb−1
Integrated luminosity:
July (EPS):
1.2 fb−1
Aug. (LP):
2.5 fb−1
Delivered: 5.6 fb−1
Recorded: 5.3 fb−1
90%–96% of recorded data usablefor analysis (depending on analysis).
Peak lumi at ATLAS: 3.6× 1033 cm−2 s−1
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 2 / 46
ATLAS detector
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 3 / 46
Pileup
Large luminosity means having many interactions in the same crossing (“pileup”)
Up to ∼ 20 — worse than expected forthis luminosity.
∼ 2(20) events/crossing forL = 1033 (1034).
50 ns bunch spacing, otherparameters pushed beyond design.
–Aug Sep–Oct
Z → µµ
Event with 20 vertices.(Ellipses enlarged 20× forvisibility.)
Challenges for reconstruction, trigger, and computing requirements.Precise modeling of pileup is essential.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 4 / 46
Trigger
Single-lepton triggers used where possible, augmented with others (eg,multi-lepton triggers) where needed.
Triggers adjusted over the course of 2011 to adjust to the increasing luminosity.
Triggers for Higgs analyses mostly stable, with a few increases of trigger threshold.
!"#"$%$&'"()
*+,-./0#)1230.0#)
4,.)1.(%)1$"))"#5&*6
!%0#)1'7238)"-)
!"#!$&6/"55,/&92,/(."0#&:;<<
Typical recorded rates: e/γ: 100 Hz; Muons: 150 Hz; jets/τ/E/T : 100 Hz
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 5 / 46
SM Higgs production and decay
HqqH
WHZHttH
pp → X + H
γγ: < 130 WW → `ν`ν: 125–180ZZ → 4`: 125–300 ZZ → ``νν: 300–600
Much progress has been made in theoretical predictions (LHC xsect WG etc)
Calculations to NNLO in most cases; uncertainties <∼ 20%
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 6 / 46
Previous ATLAS/CMS combinationCombination as of Nov., using data through Aug. 11 (∼ 2.3 fb−1 each)
LEP 95% CL:mH > 114.4 GeV
Tevatron 95% exclusion:100 < mH < 109 GeV;156 < mH < 177 GeV
Excluded at 95% CL: 141–476 GeVExcluded at 99% CL: 146-443 GeV (except a few small regions within 200-300 GeV).Expected exclusion: 124–520 GeV. Max deviation from BG-only @ 144 GeV (3σ).
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 7 / 46
Summary of ATLAS SM Higgs searches
Substantial progress over the past year in understanding the detector andin improving reconstruction and simulation.
Channel mH range Int lum Major Signal events S/B after Expected
(GeV) fb−1 bkgs after cuts cuts σ/σSMsensitivity
H → γγ 110–150 4.9 γγ, γj , jj 70 0.02 1.6–2H → ττ → `` + ν 110–140 1.1 Z → ττ , top 0.8 0.02 30–60H → ττ → `τhad 100–150 1.1 Z → ττ 10 5E-3 10-25W/ZH → bb`(`) 110–130 1.1 W/Z + jets, top 6 5E-3 15–25
H → WW (∗) → `ν`ν 110–300 2.1 WW , top, Z + jet 20 (@130) 0.3 0.3–8
H → ZZ (∗) → 4` 110–600 4.8 ZZ∗, top, Zbb 2.5 (@130) 1.5 0.7–10
H → ZZ (∗) → ``νν 200–600 2.1 ZZ , top, Z + jets 20 (@400) 0.3 0.8–4H → ZZ → ``qq 200–600 2.1 Z + jets, top 2–20 (@400) 0.05–0.5 2–6H → WW → `νqq 240–600 1.1 W + jets, top, jets 45 (@400) 1E-3 5–10
γγ and ZZ → 4` channels updated with full luminosity.WW → `ν`ν also updated.(Three channels with the best sensitivity in the allowed low-mass region.)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 8 / 46
H → WW (∗) → `ν`ν∫L dt = 2.1 fb−1
Covers 110 < mH < 300 GeV.
Relatively large σ ∼ 200 fbI Most sensitive channel over
125–180 GeV.
But 2ν: no mass reconstruction.
Initial selection: 2opposite-charge leptonsinconsistent with Z/J/ψ decay,large E/T .
0- and 1-jet bins; b-jet veto.
Additional selections to reduceWW : pT (``), m(``), ∆φ(``).
Final selection:0.75mH < mT (``,E/T ) < mH .
Backgrounds: WW , top, W /Z + jetsCompare data to background MC insignal-free control regions; derive scalefactors by which MC should be corrected.
Control MC Observedregion expectation in data
WW + 0 jets 296± 36 296WW + 1 jet 171± 21 184top 1-jet 270± 69 249
jetsN
0 2 4 6 8 10
Eve
nts
200400600800
10001200140016001800200022002400
Data stat)⊕ SM (sys
Diboson Top
Z+jets H [150 GeV]
W+jets (data driven)
ATLAS-1 L dt = 2.05 fb∫ = 7 TeV, s
νlνl→WW→H
Njets after
`, m(``), E/T cuts
Need good understanding of E/T and background.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 9 / 46
H → WW (∗) → `ν`ν
E/T spectrum in data for µµ well-described by background model.
[GeV]T,relmissE
0 20 40 60 80 100 120 140 160 180 200
Eve
nts
/ 4 G
eV
-110
1
10
210
310
410
510
610
710
810 Data stat)⊕ SM (sys
Diboson Top
Z+jets H [150 GeV]
W+jets
ATLAS-1 L dt = 2.05 fb∫ = 7 TeV, s
νµνµ→WW→H
For E/T > 50 GeV, mostly realν’s from top.
For E/T < 50 GeV, a mixture offakes and real ν’s from Z + jets.
After all cuts (for mH = 130 GeV)
Observed in data 94 events10 ee, 42 eµ, 42 µµ
Expected BG 76± 11Expected signal 19± 4
Transverse mass spectrum before mT cut:
mT =q
(E ``T + E/T )2 − (P``T + PmissT )2
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 10 / 46
H → WW (∗) → `ν`ν
Exclusion limits:
[GeV]Hm120 140 160 180 200 220 240 260 280 300
SM
σ/σ95
% C
L Li
mit
on
-110
1
10
210
ObservedExpected
σ 1 ±σ 2 ±
-1 Ldt = 2.05 fb∫ = 7 TeVs
νlνl→(*)WW→HATLAS
Consistency with background-onlyhypothesis
[GeV]Hm100 150 200 250 300
0p
-710
-610
-510
-410
-310
-210
-110
1
ObservedExpected
σ2
σ3
σ4
σ5
-1 Ldt = 2.05 fb∫ = 7 TeVs
νlνl→(*)WW→HATLAS
Excluded at 95% CL: 145 < mH < 206 GeV (expected: 134–200 GeV)Maximum deviation is 1.9σ at mH ∼ 130 GeV.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 11 / 46
H → γγ∫L dt = 4.9 fb−1
Covers 110 < mH < 150 GeV.
γγ Higgs decay relatively rareI σ × BR ∼ 40 fb
Simple final state: 2 high-pT
isolated photons.I Require
ET (γ1, γ2) > (40, 25) GeV
Events divided into 9 categoriesbased on:
I Photon η (central, crack,endcap).
I Did photon convert?I pT (γγ) perpendicular to γγ
thrust axis.
Main background: continuum γγproduction.
Other backgrounds: γj , jj .
In 4.9 fb−1, expect
70 signal events formH = 125 GeV.
∼ 3000 background events insignal window.
S/B ∼ 0.02.
Analysis requires excellent γγ mass resolution and π0 rejection.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 12 / 46
m(γγ): opening angle [m2γγ = 2E1E2(1− cosα)]
Fine EM calorimeter segmentationallows direct θ(γ) measurement.
Crucial at high pileup with manyvertices over the ∼ 5.6 cm LHCbeam spot:Which one produced the γγ pair?
2. Find vertex zX
Calorimeter “pointing” reducesvertex uncertainty: ∼ 5.6→ 1.5 cm.
With pointing (and conversion vertexlocation), angular term contributesnegligible uncertainty to m(γγ).
Robust against pileup!
z(γ1)− z(γ2)
σ(z) ∼ 1.5 cm
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 13 / 46
m(γγ): resolution
Calorimeter energy calibration:
From Z , J/ψ, W data and MC
Scale @mZ known to 0.5%.
Linearity better than 1%.
Uniformity: ∼ 1% (barrel),∼ 1.7% (endcap).
e calibration converted to γ viaMC (small systematics fromupstream material).
mee
Width of m(γγ):
Overall: 1.7 GeV.
Best: 1.4 GeV (unconvertedcentral).
Worst: 2.3 GeV (≥ 1γconverted, ≥ 1γ in crack).
[GeV]γγm110 112 114 116 118 120 122 124 126 128 130
/ 0.
5 G
eVγγ
1/N
dN
/dm
0
0.02
0.04
0.06
0.08
0.1
0.12ATLAS Preliminary
(Simulation)
=120 GeVH
, mγγ→ H→gg
Fit
6≤ µ 9≤ µ6 <
12≤ µ9 <
> 12µ
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 14 / 46
π0 rejectionImportant backgrounds are jj and γjwith j → π0.
Need for π0 rejection motivated fineη segmentation of first EM compart-ment.
(ET ∼ 20–30 GeV)
η-strips
jj : ∼ 500 µb
γj : ∼ 200 nb
γγ: ∼ 30 pb
H → γγ: ∼ 40 fb
Large uncertainties on σ(γj , jj): need to measure with data.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 15 / 46
H → γγ background composition
After all cuts: 22489 events in 100 < mγγ < 160 GeV.
Estimate composition using control samples
γγ jγ jj DYγγ jγ jj DY
Num
ber
of E
vent
s
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
Data-driven estimations
expectedγγj expectedγ
jj expected
DY expected
ATLAS Preliminary
-1 Ldt=4.9 fb∫
γγ jγ jj DYγγ jγ jj DY
Num
ber
of E
vent
s
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000 Nb. of events Fractionγγ 16000± 1120 (71± 5)%γj 5230± 890 (23± 4)%jj 1130± 600 (5± 3)%DY/Z 165± 8 (0.7± 0.1)%
Purity ∼ 70%γj + jj � γγ
Photon identification efficiency: 85± 5%(From MC, cross-checked with data.)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 16 / 46
H → γγ: results
Background in each category fitwith an exponential.
[GeV]γγm
100 110 120 130 140 150 160
Dat
a -
Bkg
mod
el
-100-50
050
100100 110 120 130 140 150 160
Eve
nts
/ 1 G
eV
0
100
200
300
400
500
600
700
800
Data 2011Background model
= 120 GeV (MC)H
SM Higgs boson m
Inclusive diphoton sample
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
Dominant systematics:
Signal yield: ∼ 20%
mγγ resolution: ∼ 14%
H → γγ pT : ∼ 8%
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 17 / 46
H → γγ: exclusion limits
[GeV]Hm
110 115 120 125 130 135 140 145 150
S
Mσ/σ
95%
CL
limit
on
1
2
3
4
5
6
7
8 limitsObserved CL limit
sExpected CL
σ 1±σ 2±
ATLAS Preliminaryγγ →H
= 7 TeVsData 2011,
-1Ldt = 4.9 fb∫
Max. deviation from background:mH = 126 GeV
Local p0: 0.27% (2.8σ)
SM Higgs expectation: ∼ 1.4σ
Include “look-elsewhere” effect:p0 (110–150 GeV): 7% (1.5σ)
Excluded at 95% CL:
114 ≤ mH ≤ 115 GeV;135 ≤ mH ≤ 136 GeV
Consistency of data withbackground-only expectation:
[GeV]Hm
110 115 120 125 130 135 140 145 150
0p
-410
-310
-210
-110
1
10 0Observed p
0 expected pγγ →SM H
= 7 TeVsData 2011,
-1Ldt = 4.9 fb∫
ATLAS Preliminary
σ1
σ2
σ3
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 18 / 46
H → ZZ (∗) → 4` (4e, 4µ, 2e2µ)∫L dt = 4.8 fb−1
Covers 110 < mH < 600
Small σ ∼ 2–5 fb. But:I Fully reconstructed mass.I Small background, S/B ∼ 1.
2 opposite-sign, same-flavorlepton pairs.
pT1,2,3,4 > 20, 20, 7, 7 GeV.
m12 within mZ ± 15 GeV
m34 > 15–60 GeV (dep on mH)
Backgrounds:
ZZ (∗) (irreducible)
At lower masses: Zbb, Z + jets,tt̄.
Reducible backgroundssuppressed with isolation andimpact parameter cuts.
Analysis requirements
High lepton reconstruction/identification efficiencies across pT spectrum.Good lepton energy/momentum resolution.Good control of reducible backgrounds for low mH .
(Need to compare against background-enriched control regions.)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 19 / 46
H → ZZ (∗) → 4` (4e, 4µ, 2e2µ): Electron performanceElectron efficiency measured withJ/ψ → ee, W → eν, Z → ee.
Crucial to use data in the low-pT
region, where material effects aremost significant.
Variation of electron efficiency withpileup well modeled by simulation:Z → ee data/MC
Systematics:
6% (pT ∼ 7 GeV)
< 2% (pT ∼ 50 GeV)
σm(H → 4e): 2.5 GeV
[GeV]eeeem
80 90 100 110 120 130 140 150
a.u.
/ 0.
5 Ge
V
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08 ATLAS Simulation Preliminary
4e→ZZ→H
0.06) GeV± = (2.53 σ
= 130 GeVHmgaussian fit
: 0.18σ 2±fraction outside
No Z mass constraint
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 20 / 46
H → ZZ (∗) → 4` (4e, 4µ, 2e2µ): Muon performance
µ reconstruction efficiency > 95%over 4 < p < 100 GeV.
Recently improved µ alignment:
MC (perfect): 2.31± 0.01 GeVData Spring 2011: 2.89± 0.01 GeVData Summer 2011: 2.45± 0.01 GeV
µ isolation eff., Z → µµ
σm(H → 4µ): 2 GeV
[GeV]µµµµm
80 90 100 110 120 130 140 150
a.u. /
0.5 G
eV
0
0.02
0.04
0.06
0.08
0.1 ATLAS Simulation Preliminary
µ4→ZZ→H
0.03) GeV± = (1.98 σ
= 130 GeVHmgaussian fit
: 0.15σ 2±fraction outside
No Z mass constraint
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 21 / 46
H → ZZ (∗) → 4` (4e, 4µ, 2e2µ): After all selectionsFull mass range
Observe 71 (24 4µ, 30 2e2µ, 17 4e)Expect from background: 62± 9
[GeV]4lm200 400 600
Eve
nts/
10 G
eV
0
2
4
6
8
10
12
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
DATABackground
=150 GeV)H
Signal (m=190 GeV)
HSignal (m
=360 GeV)H
Signal (m
PreliminaryATLAS
m(4`) < 180 GeV
Observe 8 (3 4µ, 3 2e2µ, 2 4e)Expect from background: 9.3± 1.5
[GeV]4lm100 150 200 250
Eve
nts/
5 G
eV
0
2
4
6
8
10
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
DATABackground
=125 GeV)H
Signal (m=150 GeV)
HSignal (m
=190 GeV)H
Signal (m
PreliminaryATLAS
For mH < 141 GeV (not already excluded), observe 3 events:two 2e2µ (m = 123.6 GeV, m = 124.3 GeV), one 4µ (m = 124.6 GeV)
In 117 < m(4`) < 128 GeV(around mH = 125 GeV):Expect 1.5 events from both signal and BGS/B ∼ 2 (4µ), 1 (2e2µ), 0.3 (4e)Background: ZZ (∗), Z + jets (4e only)
Main systematics:Higgs cross section: ∼ 15%Electron efficiency: ∼ 2–8%ZZ (∗) background: ∼ 15%Zbb,Z + jets backgrounds: ∼ 40%
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 22 / 46
4µ candidate with m(4µ) = 124.6 GeV
pT (µ−, µ+, µ+, µ−) = 61.2, 33.1, 17.8, 11.6 GeV; m12 = 89.7 GeV, m34 = 24.6 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 23 / 46
2e2µ candidate with m(4`) = 124.3 GeV
pT (e+, e−, µ−, µ+) = 41.5, 26.5, 24.7, 18.3 GeV; mee = 76.8 GeV, mµµ = 45.7 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 24 / 46
H → ZZ (∗) → 4` backgroundsCheck reducible backgrounds with background-enriched control samples:
2 opposite-sign, same-flavor leptons near mZ ;
2 additional same-flavor leptons passing all selections except isolation and impact parameter
[GeV]34m20 40 60 80 100
Eve
nts/
10 G
eV
0
10
20
30
40
50 DATAZZZ
bZbtt
WZSyst.Unc.
PreliminaryATLAS
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
Z + µµ
[GeV]34m20 40 60 80 100
Eve
nts/
10 G
eV
0
5
10
15
20
25
30
35
40DATAZZZ
bZbtt
WZSyst.Unc.
PreliminaryATLAS
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
Z + ee
Low-mass regions dominated by Zbb (muons) and Z + jets (electrons).MC reproduces data well (within uncertainties)Z + e/Z +µ samples used to compare selection efficiencies between data and MC.Good agreement → MC used to estimate background in signal region.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 25 / 46
H → ZZ (∗) → 4` exclusion limitsFit signal and background expectations to mass spectrum.
[GeV]Hm110 120 130 140 150 160 170 180
SM
σ/σ95
% C
L lim
it on
1
10
210
sObserved CL
sExpected CL
σ 1 ±σ 2 ±
PreliminaryATLAS 4l→(*)
ZZ→H-1Ldt = 4.8 fb∫
=7 TeVs
sObserved CL
sExpected CL
σ 1 ±σ 2 ±
[GeV]Hm200 250 300 350 400 450 500 550 600
SM
σ/σ95
% C
L lim
it on
1
10
210
sObserved CL
sExpected CL
σ 1 ±σ 2 ±
PreliminaryATLAS 4l→(*)
ZZ→H-1Ldt = 4.8 fb∫
=7 TeVs
sObserved CL
sExpected CL
σ 1 ±σ 2 ±
Excluded (95% CL): 135 < mH < 156 GeV and 181 < mH < 415 GeV(except 234–255 GeV)
Expected exclusion: 137 < mH < 158 GeV and 185 < mH < 400 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 26 / 46
H → ZZ (∗) → 4`: consistency with BG-only hypothesis
[GeV]Hm100 200 300 400 500 600
0p
-610
-510
-410
-310
-210
-110
1
ObservedExpected
σ2 σ3
PreliminaryATLAS
4l→(*) ZZ→H
-1Ldt = 4.8 fb∫=7 TeVs
[GeV]Hm110 120 130 140 150 160 170 180 190 200
0p
-610
-510
-410
-310
-210
-110
1
ObservedExpected
σ2 σ3
PreliminaryATLAS
4l→(*) ZZ→H
-1Ldt = 4.8 fb∫=7 TeVs
Maximum deviations from background-only expectations
mH (GeV) Local p0 LEE corrected Local signif Expected from SM Higgs125 1.8% ∼ 50% 2.1σ 1.4σ244 1.1% ∼ 50% 2.3σ 3.2σ500 1.4% ∼ 50% 2.2σ 1.5σ
Look-elsewhere effect (LEE) estimated over 110 < mH < 600 GeVScott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 27 / 46
Combined exclusion limits
Channels: γγ; ττ ; WW (∗) → `ν(`ν, qq); ZZ (∗) → ``(``, νν, qq)Not included: W /ZH → `bb + X
[GeV]HM100 200 300 400 500 600
SM
σ/σ95
% C
L Li
mit
on
-110
1
10
ObservedExpected
σ1 ±σ2 ± = 7 TeVs
-1 Ldt = 1.0-4.9 fb∫ATLAS Preliminary 2011 Data
CLs Limits
[GeV]HM110 115 120 125 130 135 140 145 150
SM
σ/σ95
% C
L Li
mit
on
1
10ObservedExpected
σ1 ±σ2 ± = 7 TeVs
-1 Ldt = 1.0-4.9 fb∫ATLAS Preliminary 2011 Data
CLs Limits
LEP LHC 11/11
Excluded at 95% CL
112.7 < mH < 115.5 GeV; 131 < mH < 453 GeV, except 237–251 GeV
Expected exclusion if no signal: 124.6–520 GeVExcluded at 99% CL: 133 < mH < 230 GeV; 260 < mH < 437 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 28 / 46
Consistency with background-only expectations
[GeV]HM100 200 300 400 500 600
Loca
l P-V
alu
e
-710
-610
-510
-410
-310
-210
-110
1
ObservedExpected
σ2
σ3
σ4
σ5 = 7 TeVs
-1 Ldt = 1.0-4.9 fb∫
ATLAS Preliminary 2011 Data
Biggest inconsistency with expectedbackground at mH ∼ 126 GeV
SM Higgs: Expect 2.4σ (1.4σ per channel).
[GeV]HM
110 115 120 125 130 135 140 145 150
Loca
l P-V
alue
-610
-510
-410
-310
-210
-110
1
Exp. Comb.
Obs. Comb.
4l→Exp. H
4l→Obs. H
γγ →Exp. H
γγ →Obs. H
νlν l→Exp. H
νlν l→Obs. H
ATLAS Preliminary
σ2
σ3
σ4
-1 L dt ~ 2.05-4.9 fb∫
2011 Data
Local significance of excess
1.9× 10−4 (3.6σ)γγ: 2.8σ; 4`: 2.1σ; `ν`ν: 1.4σ
Corrected for look-elsewhere effect
110–146 GeV: 2.5σ (0.6%); 110–600 GeV: 2.2σ (1.4%)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 29 / 46
Consistency with SM Higgs
[GeV]HM110 115 120 125 130 135 140 145 150
Sign
al s
treng
th
-3
-2
-1
0
1
2
3Best fit
σ1 ± = 7 TeVs
-1 Ldt = 4.9 fb∫ATLAS Preliminary γγ→H
2011 Dataγγ
[GeV]HM110 115 120 125 130 135 140 145 150
Sign
al s
treng
th
-3
-2
-1
0
1
2
3
4
5
6
Best fitσ1 ±
= 7 TeVs
-1 Ldt = 4.8 fb∫ATLAS Preliminary llll→ZZ→H
2011 Data4`
[GeV]HM110 115 120 125 130 135 140 145 150
Sign
al s
treng
th
-2
-1.5
-1
-0.5
0
0.5
1
1.5
2
2.5
Best fitσ1 ±
= 7 TeVs
-1 Ldt = 1.0-4.9 fb∫ATLAS Preliminary 2011 Data
Total
[GeV]HM110 115 120 125 130 135 140 145 150
Sign
al s
treng
th
-2
0
2
4
6
8
10
12
Best fitσ1 ±
= 7 TeVs
-1 Ldt = 2.05 fb∫ATLAS Preliminary νlνl→WW→H
2011 Data`ν`ν
Excess 2± 0.8 times larger than expected in γγ channel; within 1σ forothers and combined.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 30 / 46
Still to come
On timescale of winter conferences:
Update all channels to fullluminosity.
Relax kinematic cuts to increaseacceptance at low mH .
Better S/B discrimination:I Multivariate techniques, more
analysis bins, etc.
Ongoing improvements toobject reconstruction anddetector modeling.
Combination with CMS underdiscussion.
2012 run:
With ∼ 20 fb−1 additionalluminosity per experiment:
5σ discovery at mH ∼ 125(ATLAS alone)
∼ 3σ per channel
Combining ATLAS+CMS:5σ down to ∼ 116 GeV
Doesn’t include:
Analysis improvements.
Possible increase in energy to√s = 8 TeV. (10% gain)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 31 / 46
Conclusions
The LHC and ATLAS have performed wonderfully over the past year!
We have searched for a SM Higgs boson at the LHC using up to 4.9 fb−1
of integrated luminosity in many distinct channels over 110–600 GeV;
Small excess observed around mH ∼ 126 GeV
Local significance 3.6σ from the H → γγ (2.8σ), H → ZZ (∗) → 4`(2.1σ), H →WW (∗) → `ν`ν (1.4σ) channels.
SM Higgs expectation: 2.4σ (local). Observed excess compatiblewithin 1σ.
Significance corrected for look-elsewhere effect: ∼ 1% (2.3σ)
More data are needed to draw any definite conclusions.We’re well-positioned for the year to come!
More info: https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/CONFNOTES/ATLAS-CONF-2011-163 (+references)
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 32 / 46
Backup
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 33 / 46
H → γγ by category 1
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 2 G
eV
0
20
40
60
80
100
120
140
160
180
Data 2011Exponential fit
TtCP1: Unconverted central, low p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 1 G
eV
0
50
100
150
200
250
Data 2011Exponential fit
TtCP3: Unconverted rest, low p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 4 G
eV
0
10
20
30
40
50
60
Data 2011Exponential fit
TtCP2: Unconverted central, high p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 2 G
eV
0
10
20
30
40
50
60
70
80
90
Data 2011Exponential fit
TtCP4: Unconverted rest, high p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 34 / 46
H → γγ by category 2
[GeV]γγm
100 110 120 130 140 150 160
Even
ts / 2
GeV
0
20
40
60
80
100
120
140
Data 2011Exponential fit
TtCP5: Converted central, low p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Even
ts / 1
GeV
0
50
100
150
200
250
Data 2011Exponential fit
TtCP7: Converted rest, low p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Even
ts / 1
GeV
0
20
40
60
80
100
120
Data 2011Exponential fit
CP9: Converted transition
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 4 G
eV
0
5
10
15
20
25
30
35
Data 2011Exponential fit
TtCP6: Converted central, high p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
[GeV]γγm
100 110 120 130 140 150 160
Eve
nts
/ 2 G
eV
0
10
20
30
40
50
60
70
80
90
Data 2011Exponential fit
TtCP8: Converted rest, high p
-1 Ldt = 4.9 fb∫ = 7 TeV, s
ATLAS Preliminary
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 35 / 46
H → γγ: Background model
!Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 36 / 46
H → γγ: Previous results
June 2011: PLHC July 2011: EPS
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 37 / 46
H → γγ
thrust axis
pT
ggp
Tt
pTl
pT
g1pT
g2
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 38 / 46
H → γγ: Events per category
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 39 / 46
m(γγ): opening angle [m2γγ = 2E1E2(1− cosα)]
Fine EM calorimeter segmentationallows direct θ(γ) measurement.
Crucial at high pileup with manyvertices over the ∼ 5.6 cm LHCbeam spot:Which one produced the γγ pair?
2. Find vertex zX
Calorimeter “pointing” reducesvertex uncertainty: ∼ 5.6→ 1.5 cm.
Robust against pileup!
Without pointing, mass resolutionwould be ∼ 20% worse with > 10pileup events.
[GeV]γγm110 112 114 116 118 120 122 124 126 128 130
/ 0.
5 G
eVγγ
1/N
dN
/dm
0
0.02
0.04
0.06
0.08
0.1
0.12ATLAS Preliminary
(Simulation)*=1 mβ
=120 GeVH
, mγγ→ H→gg
Fit
truth vertex
2T
pΣ
Calo/Conv pointing
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 40 / 46
H → ZZ (∗) → 4` kinematics
[GeV]12m70 80 90 100 110
Eve
nts/
5 G
eV
0
10
20
30
40
50
60
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
DATA
(*)ZZ
t,tbZ,Zb
PreliminaryATLAS
[GeV]T
p20 40 60 80 100 120 140
Eve
nts/
5 G
eV
0
5
10
15
20
25
30
35
40
DATA
ZZ
t,tbZ,Zb
ATLAS Preliminary-1Ldt= 4.8 fb∫
= 7 TeVs
[GeV]34m20 40 60 80 100 120
Eve
nts/
10 G
eV
0
5
10
15
20
25
30
-1Ldt = 4.8 fb∫ = 7 TeVs
4l→(*)ZZ→H
DATA
(*)ZZ
t,tbZ,Zb
PreliminaryATLAS
η-2.5 -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 2.5
Eve
nts/
0.4
0
10
20
30
40
50DATAZZ
t,tbZ,Zb
ATLAS Preliminary-1Ldt= 4.8 fb∫
= 7 TeVs
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 41 / 46
H → ZZ (∗) → ``νν∫L dt = 2.1 fb−1
Covers 200 < mH < 600 GeV.
σ ∼ 20 fb larger than 4`.I But larger background.I Still most sensitive channel for
mH > 300 GeV.
Selection:I Opposite-sign ee or µµ pair
with |m`` −MZ | < 15 GeV.I No b-tagged jets.I E/T not pointing along a jet.I Further cuts on E/T and ∆φ``
dep. on hypothesized mH .
mH < 280 GeV mH > 280 GeVE/T > 66 GeV E/T > 82 GeV1 < ∆φ`` < 2.64 ∆φ`` < 2.25
∆φE/T ,``< 1
Backgrounds:ZZ : modeled from MC.
Z + jets, top: modeled from MC, checkedagainst data control samples.
Multijet BG calculated from data.
Missing ET distribution:
0 50 100 150 200 250 300
Eve
nts
/ 5 G
eV
110
210
310
410
510
610
710
0 50 100 150 200 250 300
Eve
nts
/ 5 G
eV
110
210
310
410
510
610
710 -1 L dt = 2.05 fb∫Data
Total BackgroundTopZZ,WZ,WWZOther Backgrounds
= 260 GeV)H
Signal (m = 400 GeV)
HSignal (m
ATLAS Preliminary = 7 TeVs
νν ll→ ZZ →H
[GeV]missTE
0 50 100 150 200 250 300
Dat
a / M
C
0.81
1.2
[GeV]missTE
0 50 100 150 200 250 300
Dat
a / M
C
0.81
1.2
Background modeled using alpgen. Small
disagreement at low end goes away if pythia
used instead. Difference taken as a systematic.Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 42 / 46
H → ZZ (∗) → ``νν
∆φ between leptons from Z decay:smaller for boosted Z s.
0 0.5 1 1.5 2 2.5 3
Even
ts / 0
.4 ra
d
110
210
310
410
510
610
710
810
0 0.5 1 1.5 2 2.5 3
Even
ts / 0
.4 ra
d
110
210
310
410
510
610
710
810 Total BackgroundTopZZ,WZ,WWZOther Backgrounds
= 260 GeV)H
Signal (m = 400 GeV)
HSignal (m
-1 L dt = 2.05 fb∫Data
ATLAS Preliminary = 7 TeVs
νν ll→ ZZ →H
[rad]llΦ∆0 0.5 1 1.5 2 2.5 3
Data
/ M
C
0.81
1.2
[rad]llΦ∆0 0.5 1 1.5 2 2.5 3
Data
/ M
C
0.81
1.2
Z → µµ candidatem(µµ) = 93.8 GeVpT (Z ) = 156 GeVE/T = 161 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 43 / 46
H → ZZ (∗) → ``νν
Transverse mass of ``E/T system:
[GeV]Tm
200 300 400 500 600 700
Eve
nts
/ 30
GeV
0
20
40
60
80
100
[GeV]Tm
200 300 400 500 600 700
Eve
nts
/ 30
GeV
0
20
40
60
80
100 -1 L dt = 2.05 fb∫data
Total BGTopZZ,WZ,WWZ,W
= 260 GeV)H
Signal (m
ATLAS Preliminary = 7 TeVs
νν ll→ ZZ →H
mH < 280 GeV
[GeV]Tm
200 300 400 500 600 700
Eve
nts
/ 50
GeV
0
10
20
30
40
50
[GeV]Tm
200 300 400 500 600 700
Eve
nts
/ 50
GeV
0
10
20
30
40
50-1
L dt = 2.05 fb∫data
Total BGTopZZ,WZ,WWZ,W
= 380 GeV)H
Signal (m
ATLAS Preliminary = 7 TeVs
νν ll→ ZZ →H
mH > 280 GeV
Final exclusion limit:
[GeV]Hm200 250 300 350 400 450 500 550 600
SM
σ/σ 9
5% C
L lim
it on
0
1
2
3
4
5
6
7
8 ObservedExpected
σ 1±σ 2±
PreliminaryATLAS
=7TeVs, -1Ldt=2.05fb∫ννll→ZZ→H
Excluded at 95% CL
310 < mH < 470 GeV
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 44 / 46
H → ZZ (∗) → ``qq∫L dt = 2.1 fb−1
Covers 200 < mH < 600 GeV
σ ∼ 10 fbI Large Z + jets background.
Selection:
Reconstruct Z → `` and Z → jj .
E/T < 50 GeV to reject tt̄.
For mH ≥ 300 GeV, additionalrequirements on ∆φ(``), ∆φ(jj),and pT (j).
Sample divided into two bins: twob-tags (“tagged”), and < 2 b-tags(“untagged”).
Tagged sample has ∼ 10× smallersignal but ∼ 10× larger S/B.
Tagged/untagged contribute aboutevenly to final result.
Backgrounds:ZZ : modeled from MC.
Z + jets, top: modeled from MC,normalized using data control samples.
Multijet BG calculated from data.
Dijet mass spectrum (untaggedselection):
[GeV] qq m
0 50 100 150 200 250
Eve
nts
/ 5 G
eV
0
500
1000
1500
2000
2500
3000
3500
4000
4500
[GeV] qq m
0 50 100 150 200 250
Eve
nts
/ 5 G
eV
0
500
1000
1500
2000
2500
3000
3500
4000
4500 -1 L dt=2.05 fb∫data Total BackgroundZ
100×Signal =300 GeV)
H(mOther Background
ATLAS Preliminary
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 45 / 46
H → ZZ (∗) → ``qq
m(``jj) for mH = 400 GeV
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
20
40
60
80
100
120
140
160
180
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
20
40
60
80
100
120
140
160
180
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
20
40
60
80
100
120
140
160
180 -1 L dt=2.05 fb∫data
10×Signal =400 GeV)
H(mTotal BackgroundZTopDiboson
ATLAS Preliminary
Untagged
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
0.5
1
1.5
2
2.5
3
3.5
4
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
0.5
1
1.5
2
2.5
3
3.5
4
[GeV]lljj m
100 200 300 400 500 600 700 800 900
Eve
nts
/ 25
GeV
0
0.5
1
1.5
2
2.5
3
3.5
4 -1 L dt=2.05 fb∫data
=400 GeV)H
Signal (m
Total Background
ZTop
Diboson
ATLAS Preliminary
Tagged
(Signal in untagged channel scaled up by ×10.)
Exclusion limits:
[GeV]Hm200 250 300 350 400 450 500 550 600
SM
σ/σ 9
5% C
L lim
it on
0
2
4
6
8
10
12
14
16
18
20 ObservedExpected
σ 1±σ 2±
PreliminaryATLAS
=7TeVs, -1Ldt=2.05fb∫llqq→ZZ→H
Close to excluding SM Higgs aroundmH = 300 GeV.
Scott Snyder (BNL) Status of Standard Model Higgs Searches ATLAS BNL Dec 20, 2011 46 / 46