[PPT] - Southern Methodist University February 13-15, 2017 Introduction PowerPoint Presentation

SLIDE 1

Santa Fe Jets and Heavy Flavor Workshop Vector boson production in pPb and PbPb collisions at the LHC and its impact on nCTEQ15 PDFs arxiv: 1610.02925

Florian Lyonnet

with A. Kusina, D.B. Clark, E. Godat, T. Jezo, K. Kovarik, F.I. Olness, I. Schienbein, J.Y. Yu

Southern Methodist University

February 13-15, 2017

SLIDE 2

Introduction W/Z@LHC Reweighting Conclusion

Methodology: Parametrization:

◮ e.g. xgp/A(x, Q2) = c0xc1(1 − x)c2ec3x(1 + ec4x)c5 ◮ ck → ck,0 + ck,1(1 − A−ck,2)

Global fit to various experimental data

◮ DIS ◮ DY ◮ Pion production

10

4

10

3

10

2

10

1

10

x

10

1

10 10

1

10

2

10

3

Q2 [GeV2 ]

NMC EMC SLAC BCDMS Hermes FNAL E665 DY: FNAL

10−2 ≤ x ≤ 0.7

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 2 / 20

SLIDE 3

Introduction W/Z@LHC Reweighting Conclusion

Methodology: Parametrization:

◮ e.g. xgp/A(x, Q2) = c0xc1(1 − x)c2ec3x(1 + ec4x)c5 ◮ ck → ck,0 + ck,1(1 − A−ck,2)

He/D Li/D Be/D C/D N/D Al/D Ca/D Fe/D Cu/D Kr/D Ag/D Sn/D Xe/D W/D Au/D Pb/D C/Li Ca/Li Fe/Be W/Be Be/C Al/C Ca/C Fe/C Sn/C Pb/C DAu/pp 0.0 0.5 1.0 1.5 2.0 2.5

χ2 /dof DIS DIS & DY DIS DIS & DY DIS DY DIS DY DIS

π0

32 11 3 47 28 3 26 33 27 12 2 8 2 9 3 3 7 7 28 28 14 14 21 14 111 14 32

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 2 / 20

SLIDE 4

Introduction W/Z@LHC Reweighting Conclusion

Methodology: Parametrization:

◮ e.g. xgp/A(x, Q2) = c0xc1(1 − x)c2ec3x(1 + ec4x)c5 ◮ ck → ck,0 + ck,1(1 − A−ck,2)

nCTEQ15

A total of 740 points after kinematical cuts

◮ χ2 = 587.4 ⇒ χ2/d.o.f = 0.81

Error analysis via the Hessian method

◮ 32 Eigenvalues Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 2 / 20

SLIDE 5

Introduction W/Z@LHC Reweighting Conclusion

10

3

10

2

10

1

1

x

2 4 6 8 10 12 14 16

xf p/A

g

10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

xf p/A

s

A=1 A=4 A=9 A=12 A=27 A=40 A=56 A=84 A=119 A=131 A=197 A=208 10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6

xf p/A

uv

10

3

10

2

10

1

1

x

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35

xf p/A

dv

10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

xf p/A

u

10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

xf p/A

d

10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

xf p/A

¯ u

10

3

10

2

10

1

1

x

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

xf p/A

¯ d

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 3 / 20

SLIDE 6

Introduction W/Z@LHC Reweighting Conclusion Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 4 / 20

SLIDE 7

Introduction W/Z@LHC Reweighting Conclusion

To go further: Include LHC data on pPb (and possibly PbPb) collisions

◮ LHCb: σ(Z → ℓ+ℓ−) ◮ ALICE: σ(W +(−) → ℓ+(−)ν(¯

ν))

◮ ATLAS: dσ(Z → ℓ+ℓ−)/dy, dσ(W +(−) → ℓ+(−)ν(¯

ν))/dy, Aℓ

◮ CMS: dσ(Z → ℓ+ℓ−)/dy,

dσ(W +(−) → ℓ+(−)ν(¯ ν))/dy, Aℓ, AFB

√spp = 7 TeV ⇒ sPbPb = 2.76 TeV √spp = 8 TeV ⇒ spPb = 5.02 TeV

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 5 / 20

SLIDE 8

Introduction W/Z@LHC Reweighting Conclusion

To go further: Include LHC data on pPb (and possibly PbPb) collisions

◮ LHCb: σ(Z → ℓ+ℓ−) ◮ ALICE: σ(W +(−) → ℓ+(−)ν(¯

ν))

◮ ATLAS: dσ(Z → ℓ+ℓ−)/dy, dσ(W +(−) → ℓ+(−)ν(¯

ν))/dy, Aℓ

◮ CMS: dσ(Z → ℓ+ℓ−)/dy,

dσ(W +(−) → ℓ+(−)ν(¯ ν))/dy, Aℓ, AFB

√spp = 7 TeV ⇒ sPbPb = 2.76 TeV √spp = 8 TeV ⇒ spPb = 5.02 TeV ⇒ Vector boson production in pPb and PbPb collisions at the LHC

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 5 / 20

SLIDE 9

Introduction W/Z@LHC Reweighting Conclusion

I) Introduction II) LHC pPb and PbPb data III) Impact on the nCTEQ15 fit - Reweighting Analysis IV) Conclusion

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 6 / 20

SLIDE 10

Introduction W/Z@LHC Reweighting Conclusion

Kinematics

x2 is the momentum fraction in the Lead nucleus

10−4 10−3 10−2 10−1 1

x1

10−4 10−3 10−2 10−1 1

x2

y = − 4 y = − 3 y = − 2 y = − 1 y = y = 1 y = 2 y = 3 y = 4

2 . 7 6 T e V W ± Z 5 . 2 T e V W ± Z

p

S = 8 T e V W ± Z

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 7 / 20

SLIDE 11

Introduction W/Z@LHC Reweighting Conclusion

Kinematics

x2 is the momentum fraction in the Lead nucleus

10-4 10-3 10-2 10-1 100

x2

5 4 3 2 1 1 2 3 4

y

W ± A W

ℓ

A W

FB

Z A Z

FB

ATLAS LHCb CMS ALICE Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 7 / 20

SLIDE 12

Introduction W/Z@LHC Reweighting Conclusion

Kinematics

x2 is the momentum fraction in the Lead nucleus

10-4 10-3 10-2 10-1 100

x2

5 4 3 2 1 1 2 3 4

y

W ± A W

ℓ

A W

FB

Z A Z

FB

ATLAS LHCb CMS ALICE

All predictions @NLO using custom FEWZ

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 7 / 20

SLIDE 13

Introduction W/Z@LHC Reweighting Conclusion

PbPb

0.0 0.5 1.0 1.5 2.0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

1 σtot dσ dyZ CMS Z →µµ PbPb,

ps =2.76 TeV

nCTEQ15 CT10

0.0 0.5 1.0 1.5 2.0

yZ

0.6 0.8 1.0 1.2 1.4

R

CT10: Lead nucleus made of CT10 protons

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 8 / 20

SLIDE 14

Introduction W/Z@LHC Reweighting Conclusion

PbPb

3 2 1 1 2 3 0.05 0.10 0.15 0.20 0.25 0.30

0.50 0.75 1.00 1.25 1.50

R

CT10: Lead nucleus made of CT10 protons

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 10 / 20

SLIDE 25

Introduction W/Z@LHC Reweighting Conclusion

Remarks:

Overall good description of the data

◮ Very good for y < 0 where nCTEQ15 has data. ◮ Some leverage in the y > 0 (low x) region ⇒ mainly

extrapolation

ALICE and LHCb ⇒ interesting with more data Potential constraining power from the ATLAS and CMS W + rapidity distributions. In particular where we don’t have data at the moment at x ∼ 3 × 10−3 ⇒ shadowing region

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 11 / 20

SLIDE 26

Introduction W/Z@LHC Reweighting Conclusion

Strange quark

Correlation of W + vs W − cross section with CMS (pPb) Strange is fixed in nCTEQ15 Compare 2 vs 5 flavours scenario

220 240 260 280 300 320 340 360 380 400

σ(pPb → W − → ℓ − ¯ ν) [nb]

250 300 350 400 450 500

σ(pPb → W + → ℓ + ν) [nb]

2 flavours 5 flavours nCTEQ15

CTEQ6. 1 + EPS

CT10 + EPS CT10 data

The impact of the strange is ∼ 30% Most of the difference between CT10+EPS09 and nCTEQ15 is the underlying strange contribution in CT10 Nuclear corrections very close to each other

◮ Consistently lower than

proton-proton results ⇒ shadowing sets up too early ?

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 12 / 20

SLIDE 27

Introduction W/Z@LHC Reweighting Conclusion

Strange quark

As we move from y < 0 to y > 0 ⇒ high-x to low-x

◮ I.e. from well constrained to less constrained

70 80 90 100 110 120 130

σ(pPb → W − → ℓ − ¯ ν) [nb]

60 70 80 90 100 110 120

σ(pPb → W + → ℓ + ν) [nb] y < − 1

CMS

80 100 120 140 160 180

σ(pPb → W − → ℓ − ¯ ν) [nb]

120 140 160 180 200 220 240

σ(pPb → W + → ℓ + ν) [nb] |y| < 1

2 flavours 5 flavours nCTEQ15

CTEQ6. 1 + EPS

CT10 + EPS CT10 data

30 40 50 60 70 80 90 100 110 120

σ(pPb → W − → ℓ − ¯ ν) [nb]

40 60 80 100 120 140 160 180 200

σ(pPb → W + → ℓ + ν) [nb] y > 1

At y < 0 nuclear corrections > 0 ⇒ pull results toward data As we move to higher y it seems we are little short ⇒ shadowing region Hypothetic: small negative s − ¯ s asymmetry at high-x ?

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 13 / 20

SLIDE 28

Introduction W/Z@LHC Reweighting Conclusion

I) Introduction II) LHC pPb and PbPb data III) Impact on the nCTEQ15 fit - Reweighting Analysis IV) Conclusion

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 14 / 20

SLIDE 29

Introduction W/Z@LHC Reweighting Conclusion

Reweighting Technique

We study the impact of the data set via a Baysian Reweighting Technique

General Idea

(i) Generate Replica: fk(x) = f0 +

i

1 2

f+

i − f− i

Rik

(ii) Calculate Giele-Keller weights: wk = exp − 1

2χ2 k 1 Nrep

i exp − 1

2χ2 i

(iii) Then one can estimate the weighted average and standard deviation: < O >= 1 Nrep

k

wkO(fk) δ < O >=

1

Nrep

k

wk(O(fk)− < O >)2

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 15 / 20

SLIDE 30

Introduction W/Z@LHC Reweighting Conclusion

Reweighting Technique

We study the impact of the data set via a Baysian Reweighting Technique

General Idea

(i) Generate Replica: fk(x) = f0 +

i

1 2

f+

i − f− i

Rik

(ii) Calculate Giele-Keller weights: wk = exp − 1

2χ2 k 1 Nrep

i exp − 1

2χ2 i

(iii) Then one can estimate the weighted average and standard deviation: < O >= 1 Nrep

k

wkO(fk) δ < O >=

1

Nrep

k

wk(O(fk)− < O >)2

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 15 / 20

SLIDE 31

Introduction W/Z@LHC Reweighting Conclusion

Reweighting Technique

We study the impact of the data set via a Baysian Reweighting Technique

General Idea

(i) Generate Replica: fk(x) = f0 +

i

1 2

f+

i − f− i

Rik

(ii) Calculate Giele-Keller weights: wk = exp − 1

2χ2 k 1 Nrep

i exp − 1

2χ2 i

(iii) Then one can estimate the weighted average and standard deviation: < O >= 1 Nrep

k

wkO(fk) δ < O >=

1

Nrep

k

wk(O(fk)− < O >)2

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 15 / 20

SLIDE 32

Introduction W/Z@LHC Reweighting Conclusion

All Data on pPb

Perform reweighting of the previously presented data on p − Pb collisions Using 104 replicas

1 2 3 4 5 6 7 8

weights

10 -1 10 0 10 1 10 2 10 3 10 4 Giele-Keller Weights

Reasonable weights, regular fall down distribution of weights No excessive values or irregular paterns

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 16 / 20

SLIDE 33

Introduction W/Z@LHC Reweighting Conclusion

All Data on pPb

LHCB Z FWD ATLAS WPM ASYM ALICE WM BWD ALICE WP FWD CMS WP ALICE WM FWD ATLAS WM LHCB Z BWD ALICE WP BWD ATLAS WP CMS WPM ASYM CMS Z CMS WM CMS Z AFB CMS WPM RATIO ATLAS Z 1 2 3 4 5 6

χ2/Ndata(Ndata = 102) before : 2. 43 after : 1. 88 χ2

before/Ndata

χ2

after/Ndata

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 17 / 20

SLIDE 34

Introduction W/Z@LHC Reweighting Conclusion

All Data on pPb

ATLAS reweighted distribution:

3 2 1 1 2 3 30 40 50 60 70 80 90 100 110

χ2/Ndata before : 1. 31 after : 0. 97

ATLAS W +

data before rew. after rew.

3 2 1 1 2 3

yℓ +

0.7 0.8 0.9 1.0 1.1 1.2 1.3

R

χ2/Ndata decreases from 1.31 to 0.97 As expected, distribution is pulled toward the data at y > 0 Not so much at y < 0 ⇒ limit of the parametrization?

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 18 / 20

SLIDE 35

Introduction W/Z@LHC Reweighting Conclusion

All Data on pPb

CMS Forward Backward: AFB(yℓ) = dNℓ±(+ylab) dNℓ±(−ylab)

0.0 0.5 1.0 1.5 2.0 2.5 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5 1.6

χ2/Ndata before : 4. 03 after : 1. 78

CMS AFB

data before rew. after rew.

0.0 0.5 1.0 1.5 2.0 2.5

yℓ

0.7 0.8 0.9 1.0 1.1 1.2 1.3

R

As expected, distribution is pulled toward the data χ2/Ndata decreases from 4.01 to 1.78

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 19 / 20

SLIDE 36

Introduction W/Z@LHC Reweighting Conclusion

Impact on the PDFs

up quark distribution at Q = 10 GeV

10-3 10-2 10-1 100 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

xuPb(x)

Q = 10 GeV, Nrep = 10000

before rew. after rew.

10-3 10-2 10-1 100

x

0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4

R

Impact of data still limited Remarks: Results obtained once these data in the fit may be different

◮ Might be able to open more

parameters ⇒ fit strange

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 20 / 20

SLIDE 37

Introduction W/Z@LHC Reweighting Conclusion

I) Introduction II) LHC pPb and PbPb data III) Impact on the nCTEQ15 fit - Reweighting Analysis IV) Conclusion

Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 21 / 20

SLIDE 38

Introduction W/Z@LHC Reweighting Conclusion

Conclusion

nCTEQ project is moving fast

◮ Publish new fit, nCTEQ15 with Hessian error analysis ⇒ PDFs

for any nuclei

◮ In the process of including LHC W/Z production data

First step: performed detailed study of the nCTEQ15 predictions for these observables Investigated reweighting technique to assess the power of this data

◮ Seems that the impact is limited in this setup ◮ Could still open possibilities and in particular to fit strange

quark PDF

◮ Will also be interesting to compare reweighting vs actual fits Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 22 / 20

SLIDE 39

Introduction W/Z@LHC Reweighting Conclusion Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 18 / 20

SLIDE 40

Introduction W/Z@LHC Reweighting Conclusion Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 19 / 20

SLIDE 41

Introduction W/Z@LHC Reweighting Conclusion Florian LYONNET Santa Fe Jets and Heavy Flavor Workshop February 13-15, 2017 20 / 20