Hadronically decaying massive particles, jet substructure, and measurement of the transverse momentum of the Z boson at LHC

(1)

Hadronically decaying massive particles, jet substructure, and measurement of the transverse momentum of the Z boson at LHC

Francesco De Lorenzi (Iowa State University)

on behalf of the ATLAS and CMS collaborations

(2)

Francesco De Lor enzi-ICHEP14 .

Introduction

• LHC experiments provided several measurements testing QCD

• At low p T : multiple soft-gluon radiation

• At high p T : hard-gluon emission

• The data are used to tune next-to-leading order plus parton shower Monte Carlo simulations

• Very high p T hadronically decaying W/Z

• provide a direct test of QCD calculations of gluon and quark radiation

• validate novel techniques of jet shapes and jet substructure for boson tagging and reducing the sensitivity to soft QCD and pileup

2

(3)

Outline

• Z p T distribution at 7 and 8 TeV and MC tuning

• Boosted Z→bb and W/Z → qq’

• Jet substructure for high p T W

3

(4)

Z p T : 7 TeV Results

• Z →ee and Z→µµ

• Low p

^T

:

• Good agreement with ResBos, Pythia,

Sherpa, Alpgen

• Sensitivity to tunes

• High p

^T

: quite good modeling by FEWZ, ResBos

]-1 [GeV Tq/dσ) dσ(1/

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

combined) data (e + µ

Z2

POWHEG+PYTHIA

Z2

PYTHIA

ProQ20

PYTHIA

Perugia 2011

PYTHIA

[GeV]

qT

0 5 10 15 20 25 30

dataδ(data - theory)/

-5 0 5

CMS

= 7 TeV s at L dt = 36 pb-1

∫

0 GeV 2

T>

.1, p 2

<

η|

|

]-1 [GeV Tq/dσ dσ1/

10-7

10-6

10-5

10-4

10-3

10-2

combined) data (e + µ

Z2

POWHEG+PYTHIA s) α2

O(

FEWZ s) O(α

FEWZ

[GeV]

qT 20 30 40 50 100 200 300

data / theory

1 2

CMS

= 7 TeV s at L dt = 36 pb-1

∫

0 GeV 2

T>

.1, p 2

<

η|

|

Phys.Rev. D85 (2012) 032002

4

arXiv:1406.3660

(5)

Z p T : CMS 8 TeV Results

High p

T

: good description by RESBOS (less good by POWHEG) as well as by FEWZ (but low stat. for p

T

> 100 GeV).

Low p

T

: sensitivity to the PS tune. Best description by PYTHIA Z2star;

POWHEG+PYTHIA Z2star provide only marginal description of the low p

T

data

1 10 10²

-1 [GeV/c] Tdqσ

d 1 σ

10-6

10-5

10-4

10-3

10-2

10-1

CMS Preliminary

Resbos data

>20 GeV

|<2.1, pT

|η

= 8 TeV s

-1 at 18.4 pb

∫

[GeV/c]

qT

1 10 10²

dataσ(data-MC)/ -5 0 5

[GeV/c]

qT -1 [GeV/c] T/dqσ dσ1/

0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08

) µ µ

→ data (Z

Pythia/Powheg Z2star Pythia Z2star Pythia P0 Pythia D6T Pythia 4C

[GeV/c]

qT

0 5 10 15 20 25 30

dataσ(data - theory)/

-5 0 5

CMS preliminary

=8 TeV s

-1 at 18.4 pb

∫

.1, 2

<

η|

| >20 GeV

pT

[GeV]

qT

]-1 [GeV T/dqσ dσ1/

10-7

10-6

10-5

10-4

10-3

10-2

data

Pythia/Powheg Z2star FEWZ CTEQ12NNLO

[GeV]

qT

20 30 40 50 60 70 100 200 300 400 500

data / theory

0 2 4

CMS preliminary

=8 TeV s

-1 at 18.4 pb

∫

.1, 2

<

η|

| >20 GeV

pT

CMS-PAS-SMP-12-025

5

(6)

ATLAS Z ɸ* 7 TeV result

Measurements at low p

T

(Z) are limited by experimental resolution and uncertainties on p

T

(l) scale. Use new observable ɸ*:

⇤ ⌘ = tan( acop /2) ⇥ sin(✓ ⌘ ⇤ )

cos(✓ ⌘ ⇤ ) = tanh ⇥

(⌘ + ⌘ )/2 ⇤

Scattering angle of

the leptons wrt the beam axis in the Z rest frame

acop = ⇡ (l, l)

• ɸ* is an angular variable. Angular resolution for leptons: ~0.5 mrad on φ and ~0.001 on η

• Correlated with p

T

and it probes ~ the same physics

Phys.Lett. B720 (2013) 32

6

(7)

ATLAS Parton Shower Tune

• New tunes to improve the

Monte Carlo (MC) description of kinematics of vector boson production:

• improved modeling of low p

T

vector boson => reliable acceptance of p

T

(l) cuts in W/Z measurements;

• crucial for W mass

measurement (and W/Z, DY for constraining PDFs).

Parameter Variation Range Variation Range Pythia8 tune Pythia8+Powheg tune

Primordial k_T [GeV] 1.0–2.5 0.5–2.5

ISR ↵^ISR_S (m_Z) 0.120–0.140 0.118

ISR cut-o↵ [GeV] 0.5–2.5 0.5–3.0

ISR ↵_S order LO NLO

Pythia8 base tune tune 4C tune 4C Powheg cut-o↵ [GeV²] - 4.0

Table 4. Parameter ranges and model switches used in the tuning of Pythia8 and Pythia8+Powheg described in section 10.

transverse momentum threshold p²_T,min, which is a steerable parameter in the program. Be- low p²_T,min, Powheg generates events without extra radiation and the phase space is popu- lated by Pythia8. Therefore, the upper limit of the Pythia8 parton shower should match the Powheg cut-o↵ value. The tunes are performed using p²_T,min = 4 GeV², corresponding to p^Z_T = 2 GeV. In addition, in order to avoid discontinuities in the matched spectrum, the

↵_S(m_Z) value used to calculate the QCD radiation in Powheg should match ↵_S^ISR(m_Z) in Pythia; ↵_S(m_Z) = 0.118 is used as in the CT10 PDFs. Correspondingly the running of ↵_S in the parton shower calculation is set to NLO. The tuning of Powheg+Pythia8 hence only varies the shower cut-o↵ and the primordial k_T in Pythia8. The other steerable parameters not used in the tuning are set to the values defined by the 4C tune.

The tunes are performed using the Professor [52] package, which interpolates the depen- dence of MC predictions on the model parameters as originally proposed in ref. [53]. Predic- tions for the p^Z_T distribution are generated at randomly chosen parameter settings (anchor points) in the ranges indicated in table 4. A fourth-order polynomial is used to approxi- mate the generator predictions between the anchor points. The optimal parameter values are determined using a ² minimization between the interpolated generator response and the data.

The sensitivity of the generator parameters to the p^Z_T and ^?_⌘ measurements is probed by performing tunes of Pythia8 and Powheg+Pythia8 to each measurement separately.

As shown in table 5 both measurements have comparable sensitivity and yield compatible tuned parameter values. As a further check of the compatibility between the p^Z_T and

?⌘ measurements, the p^Z_T-tuned and ^?_⌘-tuned predictions are compared to the measured p^Z_T distribution. The tuning uncertainty is obtained from variations of the eigenvector components of the parameters error matrix over a range covering ² = ²_min/dof. Figure 9 shows that the tuned predictions agree with the measured cross sections within 2% for p^Z_T < 50 GeV, and with each other within the tuned parameter uncertainties.

Since the p^Z_T and ^?_⌘ observables provide similar sensitivity to the parton shower parameters and to avoid correlations between these measurements, the final tune optimally combines

– 23 –

• Use both Z p

T

and Z ɸ* data

• Independent tunes

• MC generators studied:

• PYTHIA8, POWHEG+PYTHIA8

7

arXiv:1406.3660

(8)

High p T Z→bb

• Phase space: 2 anti-k

T

b-jets with R=0.4,

• p

^T

(jet)>40 GeV,

• |η(jet)|<2.5,

• ΔR<1.2,

• p

^T

(dijet)>200 GeV,

• 60 < M(dijet) < 160 GeV

• Separate Signal and control regions with a NN S

NN

based on Δη(dijet, balancing jet) and |η(dijet)|

• Very small correlation between NN and dijet mass

• Constrain the background shape from data in background enriched control region

8

arXiv:1404.7042

Signal

Background

(9)

High p T Z→bb

• Background: Simultaneous fit in signal and control region

• Signal model: allow peak position and yield as free parameters, other

parameters from MC

• Other small backgrounds from MC

• Acceptance from MC

Z ! bb (p dijet T > 200 GeV ) =

2.02 ± 0.2(stat.) ± 0.25(syst.) ± 0.06(lumi.)pb

POWHEG = 2.02 +0.25 0.19 (scales) +0.03 0.04 (PDF)pb

NLO ME+PS predictions:

aMC@NLO = 1.98 +0.16 0.08 (scales) ± 0.03(PDF)pb

Very good agreement with predictions

9

(10)

Boosted W/Z → qq’

• Reconstruct hadronic decay product in a single jet

• anti-k T R=0.6

• p T > 320 GeV

• Boosted back the constituents in the jet center of mass frame

• QCD jets isotropic distribution while W/Z jets back-to-back topology

• Build a discriminating

likelihood variable using jet shape variables in the jet CoM frame (thrust, sphericity and aplanarity)

10

Jet center-of-mass frame

arXiv:1407.0800

NEW!

Likelihood

-0.5 0 0.5

Jets / 0.04

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14

QCD jets W/Z jets Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

> 320 GeV pT

< 1.9

(11)

• Signal extracted with fit to the jet mass distribution

• Background model extensively studied in control regions, in different kinematic ranges and with different MC

• Signal Model from MC

• Other small background from MC

• Signal acceptance from MC

11

W+Z

(W

^±

/Z ! qq; p

_T

> 320 GeV, | ⌘ | < 1.9) = 8.5 ± 0.8 (stat.) ± 1.5 (syst.) pb.

W MCFM +Z = 5.1 ± 0.5 pb. Consistent with MCFM prediction within 2σ

Boosted W/Z → qq’

NEW!

Jet Mass [GeV]

50 100 150 200 250

Jets / 2 GeV

0 5000 10000 15000 20000 25000

Data 2011

Signal + Background fit Background fit component Signal fit component

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 320 GeV |

pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Data - Fit bkg

-200 0 200 400 600 800 1000 1200 1400 1600 1800

(12)

Francesco De Lor enzi-ICHEP14 .

Jet substructure: Boosted W

• Exploit jet substructure and jet shapes techniques for W tagging in searches and measurement

• Studies performance of different

• jet algorithm (anti-kt, CA)

• grooming combinations

(trimming, pruning, split-filter, BDRS)

• Jet R (0.8, 1.0, 1.2)

• p

^T

bins

• Data/MC comparison in QCD/

Signal enriched samples

ATL-PHYS-PUB-2014-004 CMS-PAS-JME-13-006

pruned jet mass (GeV)

0 50 100 150

Normalized Distribution

0 0.2 0.4

, Pythia6 WL

WL

X →

+ <PU> = 22 + sim.

+ <PU> = 12 + sim.

W+jets, MG+Pythia6 + <PU> = 22 + sim.

+ <PU> = 12 + sim.

= 8 TeV, W+jets s

CMS Preliminary Simulation,

CA R=0.8 < 350 GeV 250 < pT

|<2.4

|η

12

(13)

W-tagging variables

• Several tagging variables studied

• splitting scales,

width n-subjettines, mass drop, Q-jets, planar flow, subjet balance, energy

correlation functions (and MVA

combinations)

• Tagging performance optimized in p

T

bins and for different jet algorithm, grooming and size

τ1 2/ τ

0 0.2 0.4 0.6 0.8 1

Normalized Distribution

0 0.1 0.2 0.3

, Pythia6 WL

WL

X →

+ <PU> = 22 + sim.

+ <PU> = 12 + sim.

W+jets, MG+Pythia6 + <PU> = 22 + sim.

+ <PU> = 12 + sim.

CA R=0.8 < 350 GeV 250 < pT

|<2.4

|η

< 100 GeV 60 < mJ

∈sig

0 0.2 0.4 0.6 0.8 1

bkg∈1 -

0 0.2 0.4 0.6 0.8 1

Neural Network (MLP) Likelihood

τ1 2/ τ ΓQJet

pruned τ1

2/ τ

kT axes τ1

2/ τ

=1.7) (β C2

mass drop

CA R = 0.8 < 350 GeV 250 < pT

|<2.4

|η

< 100 GeV 60 < mj

13

(14)

Francesco De Lor enzi-ICHEP14 .

Conclusion

• Many measurements that probe kinematics of W/Z production at LHC.

• p T (Z): low p T gives sensitivity to the parton-shower model and to resummation effects; high p T – test of higher-order calculations

• Z ɸ*: probes same physics as p T , but doesn't depend on

momentum scale; test of resummation and parton-shower models

• MC tunes to p T /ɸ* data: improved modeling of the low-p T region (crucial for W-mass measurement.)

• Hadronically decaying boosted boson:

• Important check for searches using boosted system

• measurement in agreement with prediction (fixed order or ME +PS)

14

(15)

Backup

15

(16)

MC Generators/Calculation

16

• Low pT: (multiple soft-gluon radiation):

• resummation up to NNLL (RESBOS) with 2 different non- perturbative parameterization used to perform the resummation

• parton shower (PS) techniques (PYTHIA, HERWIG),

• ME+PS with ME O(α S ) (MC@NLO, POWHEG);

• High pT: (hard-gluon emission):

• fixed-order calculations up to O(α 2 S ) (FEWZ,

DYNNLO) V+jet @NLO: MCFM, Blackhat+Sherpa)

• multi-leg tree-level ME+PS (SHERPA, ALPGEN).

(17)

Grooming

17

(18)

18

(19)

Z→bb

19

Dominant trigger Other triggers p

T

> 250 GeV

Different kinematic regions and triggers provide consistent measurements

(20)

W/Z → qq’

20

α = pbal/Mdijet, where pbal is the transverse momentum of the best balancing jet and Mdijet is the invariant mass of the candidate jet - balancing jet system.

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 200 400 600 800

1000 ATLAS

= 7 TeV, 4.6 fb-1

s

> 320 GeV pT

< 0.2 < 1.9 PT

Control region fit

Data fit background

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 200 400 600 800 1000

1200 ATLAS

= 7 TeV, 4.6 fb-1

s

> 320 GeV pT

< 1.9 Second leading L Jet

Control region fit

Data fit background

Jets / 0.04

0 0.02 0.04 0.06 0.08 0.1

0.12 ^{Data 2011}

PYTHIA8 PYTHIA6 AMBT1 PYTHIA6 AMBT2B HERWIG++

PYTHIA6 PERUGIA2011 POWHEG + PYTHIA

ATLAS

= 7 TeV, 4.6 fb-1

s

> 320 GeV pT

< 1.9

Likelihood

-0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6

MC/Data

0.6 0.8 1 1.2 1.4

Likelihood -0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6

Jets / 0.04

0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18

PYTHIA8 PYTHIA6 AMBT1 PYTHIA6 AMBT2B HERWIG

ATLAS Simulation = 7 TeV

s

> 320 GeV pT

< 1.9

(21)

Francesco De Lor enzi-ICHEP14 .

W/Z → qq’

21

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 5000 10000 15000 20000 25000

Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 320 GeV | pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 2000 4000 6000 8000 10000 12000 14000 16000 18000 20000

22000 Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 340 GeV | pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 2000 4000 6000 8000 10000 12000 14000 16000 18000

Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 360 GeV | pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 2000 4000 6000 8000 10000 12000

14000 ^{Data 2011}

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 380 GeV | pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 2000 4000 6000 8000 10000 12000

Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 400 GeV | pT

L > 0.15

Jet Mass [GeV]

50 60 70 80 90 100 110 120 130 140

Jets / 2 GeV

0 2000 4000 6000 8000 10000

Data 2011

ATLAS

= 7 TeV, 4.6 fb-1

s

| < 1.9 > 420 GeV | pT

L > 0.15

(22)

Francesco De Lor enzi-ICHEP14 .

Data/MC comparison

• Compare data and MC in background and signal enriched samples

• Background: dijet sample

• Signal: Semileptonic ttbar events (high pT lepton/

MET OR HepTopTagged events)

0 50 100 150

Events

0 0.2 0.4 0.6 0.8

106

×

data

QCD MG+Pythia6 QCD Herwig++

QCD Pythia8 CA R=0.8

< 600 GeV 400 < pT

|<2.4 η

|

= 8 TeV, dijets s

-1, CMS Preliminary, 19.6 fb

pruned jet mass (GeV)

0 50 100 150

Data / Sim 0

1 2

dijets

ttbar

0 0.2 0.4 0.6 0.8 1

Events

0 0.2 0.4 0.6 0.8 1 1.2

106

×

data

QCD MG+Pythia6 QCD Herwig++

QCD Pythia8 CA R=0.8

< 600 GeV 400 < pT

|<2.4

|η

= 8 TeV, dijets s

-1, CMS Preliminary, 19.6 fb

τ1 2/ τ

0 0.2 0.4 0.6 0.8 1

Data / Sim 0

1 2

ttbar dijets

22

(23)

b-jet shapes in ttbar events

• Ttbar event selection

• Jet shapes comparison: b-jet and light (W->qq)

• Sensitive to underlying event, soft gluon radiation, non-

perturbative effects.

• p

^T

bins up to 150 GeV

6

100 200 300 400 500 600

(r = 0.3)Ψ1 -

0.05 0.1 0.15 0.2 0.25 0.3

jets, R = 0.6 anti-kt

| < 2.8 y

|

ATLAS

Data

PYTHIA AMBT1 PYTHIA AUET2 PYTHIA Perugia2011 Pythia 8 4C

(GeV) pT

0 100 200 300 400 500 600

Data/MC

0.60.81 1.2 1.4

0.1 0.2 0.3 0.4 0.5

(r)ρ

10-1

1

10 anti-k_t jets, R = 0.6

< 40 GeV pT

30 GeV <

| < 2.8 y

|

ATLAS

Data

r

0 0.1 0.2 0.3 0.4 0.5 0.6

Data/MC

0.8 1 1.2

Jet Shape: parton-to-jet fragmentation process

Sensitive to underlying event, soft gluon radiation, non-perturbative effects.

(r): Integrated energy density

[Phys. Rev. D 83, 052003 (2011)]

Narrower jets at higher p T Pythia AMBT1 gives

slightly narrower jets but others describe data well.

r: distance to jet axis

Jet mass and substructure are also measured. [ATLAS-CONF-2011-073]

30 GeV < p T < 600 GeV,

|y|<2.8

compared to Pythia tunes.

(and to various MC predictions:

[ATL-PHYS-PUB-2011-010] )

(r): Fractional energy density

) (

1 r

(r) vs r

1- (r=0.3) vs p T

6

100 200 300 400 500 600

(r = 0.3)Ψ1 -

0.05 0.1 0.15

0.2 0.25 0.3

jets, R = 0.6 anti-kt

| < 2.8 y

|

ATLAS

Data

(GeV) pT

0 100 200 300 400 500 600

Data/MC

0.6 0.81 1.2 1.4

0.1 0.2 0.3 0.4 0.5

(r)ρ

10-1

1

10 anti-k_t jets, R = 0.6

< 40 GeV pT

30 GeV <

| < 2.8 y

|

ATLAS

Data

r

0 0.1 0.2 0.3 0.4 0.5 0.6

Data/MC

0.8 1 1.2

Jet Shape: parton-to-jet fragmentation process

Sensitive to underlying event, soft gluon radiation, non-perturbative effects.

(r): Integrated energy density

[Phys. Rev. D 83, 052003 (2011)]

Narrower jets at higher p T Pythia AMBT1 gives

slightly narrower jets but others describe data well.

r: distance to jet axis

Jet mass and substructure are also measured. [ATLAS-CONF-2011-073]

30 GeV < p T < 600 GeV,

|y|<2.8

compared to Pythia tunes.

(and to various MC predictions:

[ATL-PHYS-PUB-2011-010] )

(r): Fractional energy density

) (

1 r

(r) vs r

1- (r=0.3) vs p T

< ⇢(r) >= 1

R · N

_jets

X

jets

p

_T

(r r/2, r + r/2) p

_T

(0, R)

< (r) > 1 N jets

X

jets

p T (0, r ) p T (0, R)

ρ(r): Fractional energy density

Ψ(r): Integrated energy density

Eur. Phys. J. C (2013) 73:2676

23