Higgs fiducial and differential cross section measurements at ATLAS - - PowerPoint PPT Presentation

SLIDE 1

Dag Gillberg
 2014-12-07 CERN

Higgs fiducial and differential cross section measurements at ATLAS

SLIDE 2

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Outline

• 1. Why measure cross sections?
• 2. Definition of fiducial volume: its acceptances and NP corrections
• 3. Overview of the measurement
• 4. Signal extraction
• 5. Correction for detector effects
• 6. Uncertainties
• 7. Physics results:
• 1. Fiducial cross sections
• 2. Differential cross sections

2

SLIDE 3

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 3

Higgs differential xsec combination 02/12/2014

michaela.queitsch-maitland@cern.ch

Why cross sections?

• Cross sections offer a direct measurement of Higgs production rates in the data with

minimal assumptions on the underlying model (‘model independent’).

• Test of the compatibility of the SM with the data.
• Can compare data to a range of different theory models now and in the future.
• The inclusive Higgs production cross section is a hot topic in the theory community
• Lot of activity to calculate the ggF Higgs production cross section to N3LO.

! ! ! ! ! !

• Differential cross sections offer a model independent way of probing the properties of the

Higgs boson.

• ‘State-of-the-art’ MC generator predictions are now at NLO accuracy in QCD, with

some steps towards NNLO.

4

Higgs differential cross section measurements

SLIDE 4

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

ggF inclusive cross section

4

baseline dFG ABNY STWZ dFMMV BBFMR BBFMR

[pb]

ggF

σ 35 40 45 50 55 60

NNLO F.O. NNLO NNLL

2

π NNLO+ NNLL'

2

π NNLO+ F.O. LO

3

• part. N

F.O. LO

3

• part. N

F.O. LO

3

• part. N

LL'

3

N

H

m = µ , /2

H

m = µ = 13 TeV, s ggF inclusive cross section, Uncertainty from largest scale-var deviation from nominal

)

H

m = m Run 1 HXSWG recommendation (dFG

= 13 TeV s = 125 GeV

H

m

No EW correction, infinite top-mass approximation = 0.1171

s

α MSTW2008nnlo68cl,

Benchmark summary from ggF XS WG

SLIDE 5

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

• Presenting ATLAS Higgs cross section measurements
• Measurements performed by extracting signal in the reference peak:


all Higgs production modes included in this peak (not only ggF)

• mH = 125.4 GeV (ATLAS measured Higgs mass), 8 TeV data only (20 fb-1)
• Only presenting the measurements in the γγ and ZZ channels


(with focus on γγ)

• Measurements are designed to be as model independent as possible
• I’m not including the recently published WW(*) fiducial cross section

measurement as part of the WW paper: https://cds.cern.ch/record/1954714

• See paper for details. The approach is a bit different from the γγ and ZZ

results I will show. For example, the expected VBF contribution is subtracted.

A few initial remarks

5

ggF

fid,0j = 27.5 +5.4 −5.3 +4.3 −3.7

= 27.5 +6.9

−6.5 fb

ggF

fid,1j = 8.4 +3.1 −3.0 ± 1.9

= 8.4 ± 3.6 fb. (stat.)(syst.)

SLIDE 6

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Cross section measured

• Binning determined by available statistics

6 For γγ and ZZ Higgs kinematics: pTH, |yH| Jet activity: Njets, pjet1 Spin & CP: cos θ* γγ only beam thrust: τjet, ∑τjet VBF: mjj, pTγγjj, ∆yjj, ∆ϕ(γγ,jj) Jet activity: 2D: pTH vs Njets bins: {0,1,≥2} jets, cos θ* vs pTH |yjet1|, pTjet2, |yjet2|, HT,jets Higgs kinematics: pTt Njets 50 GeV threshold Spin & CP: ∆ϕjj ZZ only: m34 = dilepton-mass of offshell Z Fiducial regions: γγ only VBF-enhanced:


mjj >400, ∆yjj > 2.8, ∆ϕ(γγ,jj) > 2.6

Higgs + 1 lepton:
 at least one e or µ with
 pT > 15 GeV, |η|<2.47 Higgs + ETmiss > 80 GeV Jet definition jets: anti-kt R=0.4, |y|<4.4 pT > 30 GeV

SLIDE 7

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Definition of fiducial volume

• Fiducial volume defined at truth particle level
• Particles with a mean life time longer than: c𝜐 > 10 mm
• Idea: apply same selection criteria as applied in the data analysis
• Avoid model dependent extrapolation
• “Trivial” extrapolation kept in to simplify


definition (e.g. detector “crack”)

• H→γγ: require the two photons from the Higgs


to be central: |η|<2.37, and have 
 pT ≿ 44 GeV and 32 GeV (see exact def. below)

• Reco-level: also avoid barrel-endcap transition


region: 1.37<|η|<1.52 (i.e. rely on MC for 
 fraction of MC events in this region)

7

H→ZZ* Acceptance ~63% ~50% Z→

same-flavour-


• pposite-sign-


pair (SFOS)

SLIDE 8

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Fiducial acceptance

• Fiducial acceptance as a function of Higgs pT for ggF only
• Split into kinematic acceptance and photon isolation

8

[GeV]

γ γ T

p 20 40 60 80 100 120 140 160 180 200 Acceptance factors 0.4 0.6 0.8 1 1.2 1.4

Simulation internal ATLAS

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

• Iso. Eff. Uncertainty

Iso

c

• Kin. Acc. Uncertainty

A

c

• Photon isolation requirement: 


∑ET < 14 GeV


• f particles within DR<0.4, 


mimics ATLAS photon isolation analysis selection

• Note: efficiency depend on

amount of hadronic activity

• Kinematic acceptance:


both photons central: |η|<2.37
 pTγγ/mγγ > 0.35 and 0.25

• Quite stable (~61%) vs most

variables

• Depends on the Higgs boost

along z-axis (rapidty)
 Fwd Higgs → fwd decay products H→ZZ does not apply any isolation requirements Kinematic acceptance ~50%

SLIDE 9

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Fiducial acceptance

9

jets

N

1 2 3 ≥

Acceptance factors 0.4 0.6 0.8 1 1.2 1.4

Simulation internal ATLAS

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

• Iso. Eff. Uncertainty

Iso

c

• Kin. Acc. Uncertainty

A

c

|

γ γ

y | 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 Acceptance factors 0.4 0.6 0.8 1 1.2 1.4

Simulation internal ATLAS

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

• Iso. Eff. Uncertainty

Iso

c

• Kin. Acc. Uncertainty

A

c

SLIDE 10

Comparing analytical ggF predictions with data

10

gluon fusion

• ther production modes

XH = VBF+VH+ttH Non-perturbative correction factor accounting for hadronization and 
 underlying event activity ggF cross section Kinematic acceptance for Higgs decay product to fulfil fiducial requirements Efficiency for photons to fulfil particle level isolation

(part of γγ fiducial definition
 not used for ZZ)

Branching ratio Example for H→γγ inclusive fiducial cross section, mH = 125.4 GeV LHC-XS: 19.15 pb 0.228% ~63% ~98% 1.00 ~4 fb

= 30.5 fb

Fiducial Acceptance

Fiducial cross section that’s measured in data SM prediction Note: in a differential distribution, each bin defines its own fiducial

• volume. Hence equation below holds bin-by-bin.

SLIDE 11

Comparing analytical ggF predictions with data

Our estimates of the above factors are in HEP data … and the measurements of course

http://hepdata.cedar.ac.uk/view/ins1306615

SLIDE 12

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Non perturbative correction

12

jets

N 1 2 3 4 5 6 7 8 9 10 Non-Perturbative Correction Factor 0.9 0.95 1 1.05 1.1 1.15 1.2 1.25

Pythia8 AU2 CT10 UE Herwig++ UE-EE-4-LO** Pythia6 P2011C Pythia6 P2012 Pythia6 AUET2B LO** Pythia6 AUET2B CTEQ6L1 Pythia6 AMBT2B LO** Pythia6 AMBT CTEQ6L1 Uncertainty

Intenal ATLAS

• y

0.5 1 1.5 2 Non-Perturbative Correction Factor 0.95 1 1.05 1.1 1.15 1.2

Pythia8 AU2 CT10 UE Herwig++ UE-EE-4-LO** Pythia6 P2011C Pythia6 P2012 Pythia6 AUET2B LO** Pythia6 AUET2B CTEQ6L1 Pythia6 AMBT2B LO** Pythia6 AMBT CTEQ6L1 Uncertainty

Intenal ATLAS

• [GeV]

T,jet1

p 20 40 60 80 100 120 140 Non-Perturbative Correction Factor 0.95 1 1.05 1.1 1.15 1.2 1.25

Pythia8 AU2 CT10 UE Herwig++ UE-EE-4-LO** Pythia6 P2011C Pythia6 P2012 Pythia6 AUET2B LO** Pythia6 AUET2B CTEQ6L1 Pythia6 AMBT2B LO** Pythia6 AMBT CTEQ6L1 Uncertainty

Intenal ATLAS

Bin-by-bin ratio: “particle-level”/“parton-level” “parton-level”: ME+Parton-showering “particle-level”: adds hadronization+UE (and beam-breakup)

SLIDE 13

Dag Gillberg (CERN)

Differential cross section measurement overview

13

• 1. Signal extraction
• 2. Unfold to particle level

and divide by integrated
 luminosity and bin-width

• 3. Plot and compare with 


theory

a) Spit dataset into bins of variable of interest (here 4 Njets bins) b) For each bin, extract s from a s+b fit to the mγγ spectra c) Large statistical uncertainty due to small s/b a) correction for detector effects with bin-by-bin unfolding b) convert to (“differential”) cross section by dividing by

• int. lumi (and bin-width)

a) compare to particle level prediction - i.e. no need for detector simulation b) Can also compare with analytical calculations (parton level) but then need small parton→particle level
 (NP) correction

20.3 fb-1 


(±2.8%)

correction factor 
 for detector effects

SLIDE 14

Higgs boson mass

14

New e/γ calibration (spring 2014)

• Calorimeter layers individually

calibrated with µ, e and γ (A)

• Energy response stable within 0.5%

versus time and pileup (B)

• Improved material description of the

calorimeters: inactive material constrained to 2-10%X0

• Precise MVA-based EM cluster

calibration → 10% improved H→γγ

mγγ resolution

• Data-MC agreement within (small!)

uncertainty after calibration (C) (A) (B) (C)

±0.5%

(C)

Combined Higgs mass: 125.36 ±0.37 (stat) ±0.21 (syst) ZZ-γγ compatibility: 2.0σ Final ATLAS RunI Higgs mass measurement, 1406.3827 Combined Higgs mass: 125.36 ±0.37 (stat) ±0.21 (syst) ZZ-γγ compatibility: 2.0σ Final ATLAS RunI Higgs mass measurement, 1406.3827

• Uncertainties:



 
 


• µ from H→γγ



 


• γγ-ZZ compatibility



 
 


[GeV]

sys stat

• ld

0.6 0.24 new 0.21 0.37

• ld

new 2.5 σ 2.0 σ

• ld

new

1.55±0.30 1.29±0.30

Calorimeter layer intercalibration Stability vs time & pileup Data vs MC & uncertainty

SLIDE 15

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Signal extraction γγ

15

The ATLAS calorimeters are finely segmented and can effectively distinguish between isolated photons and backgrounds like π0→γγ

All diphoton events 
 with 3-or-more jets Nice Higgs resonance peak
 seen! Background estimated by smooth fit. Main systematics from 
 photon energy resolution,
 i.e. uncertainty on width 


• f the resonance peak

SLIDE 16

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Signal extraction ZZ

16

Significantly better s/b 
 compared to γγ Irreducible ZZ from MC Normalization from NLO
 calculation.
 Reducible background 
 (jets fake one or more
 leptons) estimated from
 data in control regions In 8 TeV data
 34 data events in signal 
 window: 118-129 GeV
 After subtracting background
 → 25.1 signal events

SLIDE 17

110 120 130 140 150 160

Events / GeV 500 1000 1500 2000 2500

data fit b + s b background,

= 8 TeV s , γ γ → H → pp

• 1

dt = 20.3 fb L

∫

= 125.4 GeV

H

m > 30 GeV

jet T

p = 0,

jets

N

ATLAS

[GeV]

γ γ

m

110 120 130 140 150 160

b data -

• 100

100

110 120 130 140 150 160

Events / GeV 200 400 600 800 1000

data fit b + s b background,

= 8 TeV s , γ γ → H → pp

• 1

dt = 20.3 fb L

∫

= 125.4 GeV

H

m > 30 GeV

jet T

p = 1,

jets

N

ATLAS

[GeV]

γ γ

m

110 120 130 140 150 160

b data -

• 50

50 100

110 120 130 140 150 160

Events / GeV 50 100 150 200 250 300

data fit b + s b background,

= 8 TeV s , γ γ → H → pp

• 1

dt = 20.3 fb L

∫

= 125.4 GeV

H

m > 30 GeV

jet T

p = 2,

jets

N

ATLAS

[GeV]

γ γ

m

110 120 130 140 150 160

b data -

• 50

50

110 120 130 140 150 160

Events / GeV 20 40 60 80 100 120

data fit b + s b background,

= 8 TeV s , γ γ → H → pp

• 1

dt = 20.3 fb L

∫

= 125.4 GeV

H

m > 30 GeV

jet T

p 3, ≥

jets

N

ATLAS

[GeV]

γ γ

m

110 120 130 140 150 160

b data -

• 20

20

SLIDE 18

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Correction for detector effects

18

aka unfolding

[GeV]

γ γ T

p

20 40 60 80 100 120 140 160 180 200

i

c Correction factors,

0.4 0.5 0.6 0.7 0.8 0.9 1

Simulation ATLAS = 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L Theoretical modelling uncertainty with total uncertainty

i

c

jets

N

1 2 3 ≥

i

c Correction factors,

0.4 0.5 0.6 0.7 0.8 0.9 1

Simulation ATLAS = 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L Theoretical modelling uncertainty with total uncertainty

i

c

Defined as Nreconstructed / Nparticle-level in each bin Driven by photon reconstruction efficiency: 
 ~80% per photon → 64% probability that both photons get reconstructed Also account for bin-migration.
 → Very small effect for photon/lepton defined variables → Sizeable for jet-based observables due to JES/JER and pileup
 (see larger to the right)

SLIDE 19

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Correction for detector effects

19

97.7% 2.3% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 5.0% 91.4% 3.6% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 5.0% 91.2% 3.8% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 5.0% 90.9% 4.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 5.1% 90.6% 4.3% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 2.9% 94.8% 2.3% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 2.9% 94.6% 2.5% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 1.1% 98.9%

[GeV]

γ γ T

p Particle level

0-20 20-30 30-40 40-50 50-60 60-80 80-100 100-200

[GeV]

γ γ T

p Reconstructed level

0-20 20-30 30-40 40-50 50-60 60-80 80-100 100-200

10 20 30 40 50 60 70 80 90 100

Simulation ATLAS

SLIDE 20

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 20

96.2% 3.7% 0.1% 0.0% 16.6% 79.3% 4.0% 0.1% 3.5% 21.6% 71.5% 3.4% 1.1% 5.3% 22.8% 70.8% jets

N Particle level

1 2 3 ≥ jets

N Reconstructed level

1 2 3 ≥

10 20 30 40 50 60 70 80 90 100

Simulation ATLAS

Correction for detector effects

SLIDE 21

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Correction for detector effects

21

Similar for ZZ: higher reconstruction efficiency per lepton
 but there are 4 of them, hence slightly larger overall
 correction for dector effects

SLIDE 22

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Uncertainties

22

|

γ γ

y | 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4

fid

σ /

fid

σ ∆ Fractional uncertainty on cross section,

• 2
• 1.5
• 1
• 0.5

0.5 1 1.5 2

ATLAS

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

Luminosity Correction factor syst. ⊕ Signal extraction syst. ⊕ Statistics ⊕

jets

N

1 2 3 ≥

fid

σ /

fid

σ ∆ Fractional uncertainty on cross section,

• 1
• 0.5

0.5 1

ATLAS

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

Luminosity Correction factor syst. ⊕ Signal extraction syst. ⊕ Statistics ⊕

Completely dominated by the statistical uncertainty.
 This picture will change in Run II…

• Now. Let’s jump to the results!

SLIDE 23

H→γγ fiducial cross sections

23

30.5 fb Our measurement: 43.2 +/-9.4 +/-3.2 pb

SLIDE 24

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 24

Fiducial region Measured cross section (fb) Baseline 43.2 ± 9.4 (stat.) +3.2

−2.9 (syst.) ± 1.2 (lumi)

Njets ≥ 1 21.5 ± 5.3 (stat.) +2.4

−2.2 (syst.) ± 0.6 (lumi)

Njets ≥ 2 9.2 ± 2.8 (stat.)+1.3

−1.2 (syst.) ± 0.3 (lumi)

Njets ≥ 3 4.0 ± 1.3 (stat.) ± 0.7 (syst.) ± 0.1 (lumi) VBF-enhanced 1.68 ± 0.58 (stat.)+0.24

−0.25 (syst.) ± 0.05 (lumi)

Nleptons ≥ 1 < 0.80 Emiss

T

> 80 GeV < 0.74 Fiducial region Theoretical prediction (fb) Source Baseline 30.5 ± 3.3 LHC-XS [56] + XH 34.1 +3.6

−3.5

STWZ [98] + XH 27.2 +3.6

−3.2

Hres [102] + XH Njets ≥ 1 13.8 ± 1.7 BLPTW [105] + XH 11.7 +2.0

−2.4

JetVHeto [106]+ XH 9.3 +1.8

−1.2

Minlo HJ+ XH Njets ≥ 2 5.65 ± 0.87 BLPTW + XH 3.99 +0.56

−0.59

Minlo HJJ+ XH Njets ≥ 3 0.94 ± 0.15 Minlo HJJ+ XH VBF-enhanced 0.87 ± 0.08 Minlo HJJ+ XH Nleptons ≥ 1 0.27 ± 0.02 XH Emiss

T

> 80 GeV 0.14 ± 0.01 XH

SLIDE 25

Transverse momentum

• Differential cross sections as a function of transverse

momentum of the Higgs-like resonance compared with theory for the γγ (left) and ZZ (right) fiducial regions

Consistent with SM theory predictions p-values 0.09-0.12 (γγ) 0.16-0.30 (ZZ)

25

SLIDE 26

Jet multiplicity

26

• Number of jets (anti-kt R = 0.4) with pT > 30 GeV and |y|<4.4 produced 


in association with the Higgs-like resonance

• ≥3 jets bin for ZZ only contain 1 event

Consistent with SM theory predictions p-values 0.30-0.42 (γγ) 0.28-0.37 (ZZ)

SLIDE 27

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Higgs pT in bins of Njets

27 [GeV]

γ γ T

p 20 40 60 80 100 120 140 160 180 200 [fb/GeV] ∆ +

T

p / d

fid

σ d 0.2 0.4 0.6 0.8 1 1.2

= 0.5 fb/GeV ∆ = 0,

jets

N = 0.2 fb/GeV ∆ = 1,

jets

N = 0 ∆ 2, ≥

jets

N

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

SLIDE 28

Leading jet pT

28

Consistent with SM theory predictions p-values 0.79-0.84 (γγ) 0.26-0.33 (ZZ)

• Transverse momentum of the leading jet produced in association 


with the Higgs boson (anti-kt R = 0.4, |y|<4.4)

• The first bin contains the events with no jet with pT > 30 GeV

SLIDE 29

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Higgs rapidity

29

| [fb] y / d|

fid

σ d 10 20 30 40 50 60

ATLAS data

• syst. unc.

H X ) +

ES

HR

( H → gg = 1.15)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

|

γ γ

y | 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 data / prediction 2 4

SLIDE 30

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Spin-CP: cos θ*

30

SLIDE 31

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 31

*)| [fb] θ / d|cos(

fid

σ d 20 40 60 80 100 120 140

ATLAS data

• syst. unc.

H X ) +

ES

HR

( H → gg = 1.15)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

*)| θ |cos( 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 data / prediction 2 4 6

| [fb/rad]

jj

φ ∆ / d|

fid

σ d 2 4 6 8 10 12 14

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

| [rad]

jj

φ ∆ | 0.5 1 1.5 2 2.5 3 data / prediction 5

Spin-CP: cos θ* and Dphi(j,j)

SLIDE 32

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

VBF variables

32

| [fb]

jj

y ∆ / d|

fid

σ d

• 1

10 1

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

|

jj

y ∆ | 1 2 3 4 5 6 7 8 data / prediction 2 4

| [fb/rad]

• jj

γ γ

φ ∆ / d|

fid

σ d

• 1

10 1 10

2

10

3

10

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

| [rad]

• jj

γ γ

φ ∆

• π

|

• 2

10

• 1

10 1 data / prediction 2 4 6

SLIDE 33

MC/data ratio of mean and mode of 
 differential distributions

33

SLIDE 34

Summary

• Presented ATLAS 8 TeV γγ and ZZ differential measurements
• Can be directly compared with theory predictions: now and in the future
• (γγ, ZZ soon) Available in HEPdata and + dedicated Rivet routine
• Statistical uncertainty dominant. Expect about equal statistical precision

with full 2015 dataset (10 fb-1 @ 13 TeV). By the end of Run II expect 
 100 fb-1 and x3 smaller uncertainties

• γγ and ZZ use the same bin edges, and can be combined if one adjust for

the channel dependent a) branching ratio and b) the fiducial acceptance

• Can use measurements to constrain theory, see talk by …
• Happy birthday Florian!

34

SLIDE 35

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Scalar pT sum and second jet pT

35

[fb/GeV]

T

H /

fid

σ d

• 1

10 1

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L ≥

jets

N

[GeV]

T

H 50 100 150 200 250 data / prediction 2 4

[fb/GeV]

T

p / d

fid

σ d

• 2

10

• 1

10 1

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L 1 ≥

jets

N

[GeV]

j2 T

p 20 40 60 80 100 120 140 data / prediction 2 4 6

SLIDE 36

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Leading jet rapidity, Njets(pT>50)

36

| [fb] y / d|

fid

σ d 2 4 6 8 10 12 14 16

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 1, ≥

jets

N

|

j1

y | 0.5 1 1.5 2 2.5 3 3.5 4 data / prediction 1 2 3

[fb]

fid

σ 5 10 15 20 25 30 35 40

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 50 GeV

jet T

p

jets

N 1 2 3 ≥ data / prediction 5

SLIDE 37

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Beam-thrust variables

37

[fb/GeV]

1

τ / d

fid

σ d

• 2

10

• 1

10 1 10

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 25 GeV

jet T

p 0, ≥

jets

N

[GeV]

1

τ 10 20 30 40 50 data / prediction

• 1

1 2

[fb/GeV]

i

τ Σ / d

fid

σ d

• 2

10

• 1

10 1 10

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 25 GeV

jet T

p 0, ≥

jets

N

[GeV]

i

τ Σ 20 40 60 80 100 120 140 data / prediction 2 4

τ = mT 2 cosh y∗ , y∗ = y − yγγ, mT = q p2

T + m2,

SLIDE 38

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

More jet variables

38

| [fb] y / d|

fid

σ d 1 2 3 4 5 6 7

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

|

j2

y | 0.5 1 1.5 2 2.5 3 3.5 4 data / prediction 2 4

[fb/GeV]

T

p / d

fid

σ d

• 2

10

• 1

10

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L 2 ≥

jets

N

[GeV]

j3 T

p 20 40 60 80 100 120 140 data / prediction 5

SLIDE 39

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

More VBF variables

39

[fb/GeV]

jj

m / d

fid

σ d

• 3

10

• 2

10

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

[GeV]

jj

m 100 200 300 400 500 600 700 800 900 1000 data / prediction 2 4 6

[fb/GeV]

T

p / d

fid

σ d

• 3

10

• 2

10

• 1

10 1

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJJ+P

( H → gg = 1.10)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p 2, ≥

jets

N

[GeV]

jj γ γ T

p 20 40 60 80 100 120 140 data / prediction 5 10

SLIDE 40

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

Exclusive jet pT

40

[fb/GeV]

T

p / d

fid

σ d 0.1 0.2 0.3 0.4 0.5 0.6

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L > 30 GeV

jet T

p = 1,

jets

N

[GeV]

j1 T

p 40 60 80 100 120 140 data / prediction 1 2

*)| θ |cos( 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 [fb] ∆ *)| + θ / d|cos(

fid

σ d 20 40 60 80 100 120 140 160 180 200

= 100 fb ∆ < 200 GeV,

γ γ T

p ≤ 80 = 0 ∆ < 80 GeV,

γ γ T

p

ATLAS data

• syst. unc.

H X ) +

8

Y

MiNLO HJ+P

( H → gg = 1.54)

ggF

K ( H t t + VH = VBF + H X

= 8 TeV s , γ γ → H

∫

• 1

dt = 20.3 fb L

SLIDE 41

• Measurement of fiducial and differential cross sections are corrected for detector effects

and designed to be as model independent as possible

• Corrected measured distributions can be
• direct comparison with theory (without the need of detector simulation)
• used to probe a variety of physics: fiducial cross section; kinematic properties; QCD;

associated jet activity; spin/CP; BSM Higgs scenarios …

• Fiducial definitions chosen to closely replicate analysis


selection to minimize model dependence:

• H→γγ inclusive cross section: nsig = 570±130, ci = 0.65±0.02:

• H→ZZ inclusive cross section:

Fiducial differential cross sections

41

H→γγ two isolated photons:

• pTγ1 / mγγ > 0.35, pTγ2 / mγγ > 0.25
• |η|<2.37
• isolation criteria:


ET < 14 GeV of particles in ΔR<0.4

H→ZZ 4e, 4µ or eeµµ

• e: pT > 7 GeV, |η|<2.47
• µ: pT > 6 GeV, |η|<2.7

differential cross section of bin i

number of extracted signal events 20.3 fb-1 (±2.8%) correction factor 
 for detector effects bin width

SLIDE 42

Example mγγ spectra for an Njets bin

42

SLIDE 43

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements

ggF inclusive cross sections

43

baseline dFG ABNY STWZ dFMMV BBFMR BBFMR

[pb]

ggF

σ 35 40 45 50 55 60

NNLO F.O. NNLO NNLL

2

π NNLO+ NNLL'

2

π NNLO+ F.O. LO

3

• part. N

F.O. LO

3

• part. N

F.O. LO

3

• part. N

LL'

3

N

H

m = µ , /2

H

m = µ = 13 TeV, s ggF inclusive cross section, Uncertainty from largest scale-var deviation from nominal

)

H

m = m Run 1 HXSWG recommendation (dFG

= 13 TeV s = 125 GeV

H

m

No EW correction, infinite top-mass approximation = 0.1171

s

α MSTW2008nnlo68cl,

SLIDE 44

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 44

baseline dFG ABNY STWZ dFMMV BBFMR BBFMR

[pb]

ggF

σ 35 40 45 50 55 60

NNLO F.O. NNLO NNLL

2

π NNLO+ NNLL'

2

π NNLO+ F.O. LO

3

• part. N

F.O. LO

3

• part. N

F.O. LO

3

• part. N

LL'

3

N

H

m = µ , /2

H

m = µ = 13 TeV, s ggF inclusive cross section, Uncertainty from largest scale-var deviation from nominal variation

R

µ

)

H

m = m Run 1 HXSWG recommendation (dFG

= 13 TeV s = 125 GeV

H

m

No EW correction, infinite top-mass approximation = 0.1171

s

α MSTW2008nnlo68cl,

ggF inclusive cross sections

SLIDE 45

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 45

baseline dFG ABNY STWZ dFMMV BBFMR BBFMR

[pb]

ggF

σ 35 40 45 50 55 60

NNLO F.O. NNLO NNLL

2

π NNLO+ NNLL'

2

π NNLO+ F.O. LO

3

• part. N

F.O. LO

3

• part. N

F.O. LO

3

• part. N

LL'

3

N

H

m = µ , /2

H

m = µ = 13 TeV, s ggF inclusive cross section, Uncertainty from largest scale-var deviation from nominal simultanious variation

F

µ +

R

µ

)

H

m = m Run 1 HXSWG recommendation (dFG

= 13 TeV s = 125 GeV

H

m

No EW correction, infinite top-mass approximation = 0.1171

s

α MSTW2008nnlo68cl,

ggF inclusive cross sections

SLIDE 46

Dag Gillberg (CERN) 2014-12-07 Higgs cross section measurements 46

baseline dFG ABNY STWZ dFMMV BBFMR BBFMR

[pb]

ggF

σ 35 40 45 50 55 60

NNLO F.O. NNLO NNLL

2

π NNLO+ NNLL'

2

π NNLO+ F.O. LO

3

• part. N

F.O. LO

3

• part. N

F.O. LO

3

• part. N

LL'

3

N

H

m = µ , /2

H

m = µ = 13 TeV, s ggF inclusive cross section, Uncertainty from largest scale-var deviation from nominal variations by a factor of 2

F

µ +

R

µ ,

F

µ ,

R

µ

)

H

m = m Run 1 HXSWG recommendation (dFG

= 13 TeV s = 125 GeV

H

m

No EW correction, infinite top-mass approximation = 0.1171

s

α MSTW2008nnlo68cl,

ggF inclusive cross sections