jet results from cms - wiki.bnl.gov · pdf fileparticle flow jets!from identiﬁed particles...

Jet Results from CMS

Cosmin DragoiuUniversity of Illinois at Chicago

(for the CMS Collaboration)

XIX International Workshop on Deep-Inelastic Scatteringand Related Subjects, Newport News 2011

Cosmin Dragoiu DIS 2011

! The CMS detector

! Jet reconstruction

! Jet energy calibration

! Jet energy resolution

! Inclusive jet cross section

! Dijet mass cross section

! Dijet azimuthal decorrelations

! Dijet angular distributions

Outline

2

More about jet performance and multijet measurements at CMS

in Joanna!s talk!


CMS Detector

Pixel (|!| < 2.5)3 layers (barrel)2 disks (endcap)

Tracking System (|!| < 2.5)10 layers (barrel): TIB (4L) + TOB (6L)11 disks (endcap): TID (3D) + TEC (9D)

Muon System (|!| < 2.4)4 muon stationsDT + RPC (barrel)CSC + RPC (endcap)

Hadronic CalorimeterHB + HO (|!| < 1.3)HE (1.3 < |!| < 3.0)HF (3.0 < |!| < 5.2)

Electromagnetic CalorimeterEB (|!| < 1.48) + EE (1.48 < |!| < 3.0)Preshower (1.65 < |!| < 2.6)

4T Magnet

Return Yoke CASTOR (5.2 < |!| < 6.6)

ZDC (|!| > 8.3)


CMS 2010 DATA

! Delivered by LHC: 47/pb

! Recorded by CMS: 43/pb

! Data taking efficiency > 90%! with all subdetectors

running > 85%

! Luminosity uncertainty: 4%


Jet Reconstruction

Calorimeter Jets

! from calorimeter towersTrack Jets

! from tracks

Jet plus Track

! from calorimeter towers corrected using tracker information

Particle Flow Jets

! from identified particles using all detector components

! Jet reconstruction algorithms available at CMS:

! kT, Anti-kT

! CMS default: Anti-kT R = 0.5 & 0.7


Jet Energy Calibration

6

Residual Correction(" & pT)

MC Correction(" & pT)

Offset Correction

Reconstructed Jet

Calibrated Jet

! removes pile-up and noise contributions

! flattens the jet response in ! and corrects the jet pT to particle level

! accounts for the differences between data and MC (dijet pT balance, MPF method)

(GeV)!

Tp

20 30 40 50 100 200

Dat

a / M

C

0.96

0.98

1

1.02

1.04

1.06

1.08

1.1

(GeV)!

Tp

20 30 40 50 100 200

Dat

a / M

C

0.96

0.98

1

1.02

1.04

1.06

1.08

1.1

pT balanceMPF

/ NDF = 14.2 / 122"

MC scaled for FSR and QCD bkg

anti-kT 0.5 PF

= 7 TeVs-1CMS preliminary, 2.9 pb

(GeV)T

p20 30 100 200 1000 2000

Abso

lute

sca

le u

ncer

tain

ty [%

]0123456789

10Total uncert.Total MPFPhoton scaleExtrapolationOffset (+1PU)Residuals

particle flow jets 0.5 & 0.7TAnti-k

= 7 TeVs-1CMS preliminary, 2.9 pb

(GeV)T

p20 30 100 200 1000 2000

Abso

lute

sca

le u

ncer

tain

ty [%

]0123456789

10

CMS-PAS-JME-10-010


! Determined using the dijet asymmetry method (dijet pT balancing)

! The asymmetry is:

! The jet pT resolution is related to the width of the asymmetry distribution

Jet Energy Resolution

7

[GeV]T

p50 100 200

reso

lutio

nT

jet p

0

0.1

0.2

0.3total systematic uncertainty

MC truth (c-term added)

MC truth

data

50 100 2000

0.1

0.2



MC truth

data

total systematic uncertainty


MC truth

data



MC truth

data

-1=7 TeV, L=35.9 pbs CMS preliminary 2010CaloJets

R=0.5)T

(Anti-k 0.5!| "0 < |

[GeV]T

p50 100 200

reso

lutio

nT

jet p

0

0.1

0.2



MC truth

data

50 100 2000

0.1

0.2



MC truth

data



MC truth

data



MC truth

data

-1=7 TeV, L=35.9 pbs CMS preliminary 2010PFJets

R=0.5)T

(Anti-k 0.5!| "0 < |

A =pjet1

T − pjet2T

pjet1T + pjet2

T

σ(pT )pT

=√

2σA

jet1, jet2 - leading jets in the event (3rd jet pT ! 0)

Calorimeter Jets Particle Flow JetsBetter resolution!

CMS-PAS-JME-10-014


Inclusive Jet Production I

8

! Represents an important test of the Standard Model

! Extends up to pT of 1.1 TeV and as low as pT of 18 GeV when using particle flow jets

! NLO pQCD using NLOJET++! Non-perturbative corrections (< 30%)

from PYTHIA6 and HERWIG++

! The data is corrected to particle level using an ansatz method

(GeV)T

p20 30 100 200 1000

dy (p

b/G

eV)

T/d

p!2 d

-1101

10210310410510

610710810

91010101110 3125)"|y|<0.5 (

625)"|y|<1 (#0.5125)"|y|<1.5 (#125)"|y|<2 (#1.55)"|y|<2.5 (#2

|y|<3#2.5

NP theory$NLOExp. uncertainty

R=0.5TAnti-k

Data for: = 7 TeVs-1CMS, 34 pb

(GeV)T

p20 30 100 200 1000

dy (p

b/G

eV)

T/d

p!2 d

-1101

10210310410510

610710810

91010101110

"r = "f = pTPDF4LHC

CMS-PAS-QCD-10-011


Inclusive Jet Production II

9

(GeV)T

p20 30 100 200 1000

Cor

rect

ion

0.50.550.6

0.650.7

0.750.8

0.850.9

0.951

|y| < 0.5 |y| < 1!0.5

|y| < 1.5!1 |y| < 2!1.5

|y| < 2.5!2 |y| < 3!2.5

R=0.5TAnti-k

= 7 TeVs-1CMS, 34 pb

! Main experimental uncertainties:

! Jet energy scale (3% to 5%)! Jet energy resolution (10% to 30%)

! Main theoretical uncertainties:! Scale ("r,"f) (5% to 10%)! PDF using PDF4LHC recipe (10%)

The data is compatible with NLO predictions!

(GeV)T

p20 30 100 200 1000

NP (P

DF4L

HC)

!Da

ta /

NLO

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

|y| < 0.5

PFCalo

NP theory!Data / NLOTheory uncertaintyExp. uncertainty

(GeV)T

p20 30 100 200 1000

NP (P

DF4L

HC)

!Da

ta /

NLO

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

= 7 TeVs-1CMS, 34 pb

Corrections to particle level


! Another important test of the Standard Model

! The dijet invariant mass probes the proton momentum fractions of the scattering partons:

! The data is corrected to particle level! Unfolding factors (0.95 to 0.98)

! NLO pQCD using NLOJET++

! Non-perturbative corrections (5% to 30%) from PYTHIA6 and HERWIG++

Dijet Mass Production I

10

(TeV)JJM0.2 0.3 1 2 3 4

(pb/

TeV)

max

d|y|

JJ/d

M!2 d

-310

210

710

1210

1710 < 0.5max|y|)1 10" < 1.0 (max0.5 < |y|)2 10" < 1.5 (max1.0 < |y|)3 10" < 2.0 (max1.5 < |y|)4 10" < 2.5 (max2.0 < |y|

Non Pert. Corr.#pQCD at NLO

PDF4LHCaveT

= pRµ =

Fµ

-1 = 36 pbintL

= 7 TeVs R = 0.7Tanti-k

CMS

M2JJ = x1 · x2 · s

8 · 10−4 ≤ x1 · x2 ≤ 0.25

Using particle flow jets!

CMS-PAS-QCD-10-025


! Main experimental uncertainties:

! Jet energy scale (15% to 60%)! Jet energy resolution (1%)

! Main theoretical uncertainties:! PDF using PDF4LHC recipe

(5% to 30%)! Non-perturbative correction

(2% to 15%)! Scale ("r,"f) (2% to 32%)

Dijet Mass Production II

11

< 0.5max0.0 < |y|

(TeV)JJM0.2 0.3 0.4 1 2

Dat

a/Th

eory

0.5

1

1.5 < 1.0max0.5 < |y|

(TeV)JJM0.2 0.3 0.4 1 2 3

Dat

a/Th

eory

0.5

1

1.5

< 1.5max1.0 < |y|

(TeV)JJM0.4 0.5 1 2 3

Dat

a/Th

eory

0.5

1

1.5 < 2.0max1.5 < |y|

(TeV)JJM0.6 1 2 3

Dat

a/Th

eory

0.5

1

1.5

< 2.5max

2.0 < |y|

(TeV)JJM0.7 1 2 3

Dat

a/Th

eory

0.5

1

1.5 CMS-1 = 36 pbintL

= 7 TeVs R = 0.7Tanti-k

Data/TheoryExp. Uncertainty

UncertaintyS! "PDF Scale UncertaintyNon. Pert. Uncertainty

Good agreement between data and NLO predictions!


!" Decorrelations I

12

∆ϕdijet = |ϕjet1 − ϕjet2|

! Measurement sensitive to higher order radiation without the need to reconstruct the radiated jets

! Systematic uncertainty (3% to 11%):! Jet energy scale (1% to 5%)! Jet energy resolution (1% to 5%)! Unfolding (1.5% to 8%)! Smearing (2.5%)

!Reasonable agreement between PYTHIA6, HERWIG++ and data

! MADGRAPH (PYTHIA8) predicts less (more) decorrelations than data

[ rad ] dijet

ϕ∆

]-1

[ ra

d

dije

tϕ

∆d

dije

tσ

d

dije

tσ

1

-210

-110

1

10

210

310

410

510 )410× > 300 GeV (

T

max p

)3

10× < 300 GeV (T

max 200 < p

)210× < 200 GeV (T

max 140 < p

10)× < 140 GeV (T

max 110 < p

< 110 GeVT

max 80 < p

PYTHIA6 D6T

PYTHIA6 Z2

PYTHIA8

HERWIG++

MADGRAPH

-1L = 2.9 pb

= 7 TeVs

|y| < 1.1

CMS

/2π /3π2 /6π5 π


PRL 106, 122003 (2011)


0.5

1

1.5

2

0.5

1

1.5

2 CMSCMSCMSCMSCMS > 300 GeVT

max p

0.5

1

1.5

2

0.5

1

1.5

2 < 300 GeVT

max 200 < p

0.5

1

1.5

2

0.5

1

1.5

2 < 200 GeVT

max 140 < p

0.5

1

1.5

2

0.5

1

1.5

2 < 140 GeVT

max 110 < p

0.5

1

1.5

2

0.5

1

1.5

2 < 110 GeVT

max 80 < p

PYTHIA6 D6T

PYTHIA6 Z2

PYTHIA8

HERWIG++

MADGRAPH

Systematic Uncert.

= 7 TeV |y| < 1.1s -1L = 2.9 pb

/2π /3π2 /6π5 π

[ rad ] dijet

ϕ∆

dije

tϕ

∆d

dije

tσ

d

dije

tσ

1

M

CD

AT

A

!" Decorrelations I

13

∆ϕdijet = |ϕjet1 − ϕjet2|

! Measurement sensitive to higher order radiation without the need to reconstruct the radiated jets

! Systematic uncertainty (3% to 11%):! Jet energy scale (1% to 5%)! Jet energy resolution (1% to 5%)! Unfolding (1.5% to 8%)! Smearing (2.5%)

!Reasonable agreement between PYTHIA6, HERWIG++ and data

! MADGRAPH (PYTHIA8) predicts less (more) decorrelations than data


PRL 106, 122003 (2011)


1

2

3

1

2

3CMS

NLO QCD Predictions CTEQ 6.6max

T = p

fµ =

rµ

> 300 GeVT

max p

1

2

3

1

2

3 < 300 GeV

T

max 200 < p

1

2

3

1

2

3 < 200 GeV

T

max 140 < p

1

2

3

1

2

3 < 140 GeV

T

max 110 < p

1

2

3

1

2

3 < 110 GeV

T

max 80 < p

Scale Dependencef

µ, r

µ

PDF Uncertainty

Non-Pert. Uncertainty

= 7 TeV |y| < 1.1s -1L = 2.9 pb

/2π /3π2 /6π5 π

[ rad ] dijet

ϕ∆

dije

tϕ

∆d

dije

tσ

d

dije

tσ

1

T

HE

OR

YD

AT

A

! NLO pQCD predictions using NLOJET++

! Non-perturbative corrections derived from PYTHIA6 and HERWIG++ (4% to 13%)

! Theoretical uncertainty:

! Scale ("r,"f) (< 50%)! PDF using CTEQ6.6 (2% to 9%)! Non-perturbative corrections (2% to 6%)

! NLO pQCD predictions mostly agree with data but undershoot the data for "# < 2$/3 (effectively 2 ! 4 LO)

!" Decorrelations II


1

2

3

1

2

3CMSCMSCMSCMSCMS > 300 GeV

T

max p

1

2

3

1

2

3 < 300 GeV

T

max 200 < p

1

2

3

1

2

3 < 200 GeV

T

max 140 < p

1

2

3

1

2

3 < 140 GeV

T

max 110 < p

1

2

3

1

2

3 < 110 GeV

T

max 80 < p

=1.0ISR

k=2.0

ISRk

=2.5ISR

k

=3.0ISR

k=4.0

ISRk

Systematic Uncert.

= 7 TeV |y| < 1.1s -1L = 2.9 pb

/2π /3π2 /6π5 π

[ rad ] dijet

ϕ∆

dije

tϕ

∆d

dije

tσ

d

dije

tσ

1

P

YT

HIA

6D

AT

A

! Variation of ISR parameter PARP(67) in PYTHIA6! D6T value 2.5 (tuned to D0 data)! Change by ± 0.5 and ± 1.5

!Sensitive to initial state radiation and mostly insensitive to final state radiation

! Changing PARP(67) by ± 0.5 ! ± 30% change in !"

!Could be used to further tune ISR

!" Decorrelations III


Dijet Angular Distributions I

16

! Probes the parton-parton scattering angle

! QCD predicts a relatively flat % distribution while new physics (quark compositeness) is expected to produce an excess at low values of %

! The data is corrected to particle level (< 3%)

! NLO pQCD predictions using NLOJET++! Non-perturbative corrections from PYTHIA6

and HERWIG++

! Good agreement with pQCD predictions! dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

> 2.2 TeVjjM (+0.5)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 2.2 TeVjjM1.8 < (+0.4)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 1.8 TeVjjM1.4 < (+0.3)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 1.4 TeVjjM1.1 < (+0.25)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 1.1 TeVjjM0.85 < (+0.2)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 0.85 TeVjjM0.65 < (+0.15)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 0.65 TeVjjM0.5 < (+0.1)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 0.5 TeVjjM0.35 < (+0.05)

dijet!

2 4 6 8 10 12 14 16di

jet

!/d

dije

t"

ddi

jet

"1/

0.1

0.2

0.3

0.4

0.5

0.6

0.7

< 0.35 TeVjjM0.25 <

CMS = 7 TeVs

-1L = 36 pb

DataQCD prediction

= 5 TeV+# = 5 TeV-#

#r = #f = pTave

χdijet = e|y1−y2| ∼ 1 + |cos θ∗|1− |cos θ∗|

CMS-PAS-QCD-10-016


Dijet Angular Distributions II

17

! Experimental uncertainties (< 3%):

! Jet energy scale (< 2.5%)! Jet energy resolution (< 1%)! Unfolding (< 1%)

! Theoretical uncertainties (< 9%):

! Scale ("r,"f) (< 9%)! PDF using CTEQ6.6 (< 0.5%)! Non-perturbative correction (< 4%)

! Limits obtained using a modified CLs approach:! Exclude $+ < 5.6 TeV at 95% CL

(expected $+ < 5.0 TeV )! Exclude $- < 6.7 TeV at 95% CL

(expected $- < 5.8 TeV )

Contact Interaction

!q q

qq

[TeV]+!

1 2 3 4 5 6

SC

L

0.04

0.06

0.08

0.1CMS = 7 TeVs

-1L = 36 pb

observed expected

" 1±expected " 2±expected


Summary

18

! LHC and CMS performed extremely well in 2010

! CMS has a rich jet physics program including precise QCD measurements and searches of new phenomena

! Many results are already submitted for publication or published

! Analyses are already beginning to exceed the Tevatron reach

! High probability of reaching 1/fb in the coming months (25/pb already collected)

! Many new and interesting physics results are on the way

! All CMS public results can be accessed at:

https://twiki.cern.ch/twiki/bin/view/CMSPublic/PhysicsResults

jet results from cms - wiki.bnl.gov · pdf fileparticle flow jets!from identiﬁed particles...

Documents