Dipanwita Dutta (ppt)

plantationscarfΤεχνίτη Νοημοσύνη και Ρομποτική

25 Νοε 2013 (πριν από 3 χρόνια και 6 μήνες)

70 εμφανίσεις


Jet R
AA
Studies with
Pb
-
Pb

data
@2.76
TeV

Dipanwita

Dutta

Nuclear Physics Division , BARC

Jets in CMS: Excellent probe

Jet Quenching

QGP

Direct jet quenching @
CMS

R
AA

over two decades of P
T

D.
Dutta

, India CMS Meet

2

Jet
quenching

via
large
dijet

energy

imbalance


To study the Nuclear Modification factor for inclusive jets






Gives Information: Medium modification of jets, jet broadening,


Initial state effect


Challenges and interest:

Jet unfolding
: subtraction of huge HI background, underlying
events (UE) produced simultaneously with hard nucleon
-
nucleon collision
-

properly subtracted to get “true”
momentum

Jet identification:
implementation of proper Jet quality criteria
to remove fake jets arising from noise, important in HI


Motivation

R
A
A

d
2
N
A
A
/
d
p
T
d

T
A
A
d
2

p
p
/
d
p
T
d

D.
Dutta

, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

T
A
A

N
c
o
l
l
/

p
p
i
n
e
l
R
AA

> 1 (enhancement)

R
AA

= 1 (no medium effect)

R
AA

< 1 (suppression)

3

Binary scaling

pp reference


Jet analysis : R
AA
twiki

page

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

4



Discussed with Jet R
AA
Analysis group:



Gunther

Roland (former HI spokes person, MIT),
Christof

Roland (MIT), Yen
-
Jie

Lee (MIT)
-

(co
-
ordinator

of high
p
T

group), Mathew
Ngyuan
, Marguerite
Tonjes

(UMD), Ying Lu (UMD)


Twiki

page



https://twiki.cern.ch/twiki/bin/viewauth/CMS/HIJetRAA



Some
i
nitial plan is discussed and mentioned in R
AA

page


Pb


Pb

:



Make raw and corrected jet
p
T

spectrum for iC5pu (
7/19/11.pdf
)
--
Dipanwita

Dutta

and akpu3pf



Find where
RecoJetpT

of two algorithms is fully efficient for trigger



Check Jet energy scale, jet response, jet resolution for both algorithms



Jet unfolding



Study
JetID

and Include
JetID

in analyzer




Both pp and
Pb
-
Pb

analysis will use same analyzer



Presenting results and discussion in high
p
T

meetings (EVO)



Jet analysis



Data
:

Pb
-
Pb

Run Nov.

Dec. 2010 and pp Run 2011 @

s=2.76
TeV




Simulations :
Pythia

jets embedded +
Hydjet

and
Pythia

QCD
Dijets

(to be done)



Integrated Luminosity:
Pb
-
Pb

: 7.2

b
-
1

and pp : 231 nb
-
1



Jet Trigger:
HLT from calorimeter towers,



Pb
-
Pb
: Trigger threshold
p
T
=35
GeV
/c



D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

5


1. Calorimeter jets


Calorimeter based Jet Finder
(IC5)



Iterative Cone Algorithm



R = 0.5


Tag:

icpu5calo




2. Particle Flow Jets


Particle Flow Jet Finder (PF)



Anti
-
kT

Clustering
Algorithm



R = 0.3

Tag:

akpu3pf

6


Jet analysis

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

6



Jet Trees are
reconstructed from RECO data (PAT on fly)


(
used by HI group for Dijet analysis )
by the working analyzer


Data:
hidata
/HIRun2010/
HIAllPhysics
/RECO/


Analyzer involve :
(1)
Reco
-
Extra (Heavy ion track reconstruction)


(2) Heavy ion background subtraction


(3) PAT
-
Jet and Ana
-
step


Tree Variables are :

vx
,
vy
,
vz
:
vertex position

n,
jtpt
,
corrpt
,
jteta
,
jtphi
: inclusive jet kinematics for each event

refpt
,
refeta
,
refphi
:

matched
genjet

kinematics

ntrack
,
trackpt
,
tracketa
,
trackphi
:

track kinematics

nPFcand
,
candId
,
candpt
,
candeta
,
candphi
:
particle flow candidate kinematics

(
pfid
: 1 = charged hadrons, 4 = photons)



For inclusive jet R
AA
study :

Low
p
T

jets, need sophisticated back
-
ground subtraction and jet identification:
new PAT analyzer to
be introduced.


7

7

D.
Dutta

, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

7


Jet analysis: Vertex z
-
distribution

No. of
Pb
-
Pb

events = 20743

Pb
-
Pb

Nov.
-

Dec . Run 2010 ,

s
NN
=
2.76
TeV

8


Jet analysis: P
T

distribution

8

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

8

HLT: Jet35U

Cut

<2

9

9

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

9


Jet analysis:


慮搠


摩d瑲楢畴楯t

Cut

<2

Black
-

icpu5calo jets

Red
-

akpu3pf jets

10


Jet analysis:






摩d瑲楢畴楯t

10

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

10

Cut

<2

icpu5calo jets

akpu3pf jets

11


Jet analysis: d


慮搠a


摩d瑲楢畴楯t

11

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

11

Cut

<2

icpu5calo jets

akpu3pf jets

blue
-

without cut;
black
-

leading jet
p
T

>100
GeV
/c


-

subleading

jet
p
T
>40
GeV
/c

12


Outlook

12

12

D.
Dutta
, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

12



Jet RAA analysis




Unfolding of jets




Jet Identification


Analysis group :


Pawan

Netrakanti


Dipak

Mishra

L. M. Pant

A. K.
Mohanty


R
AA

analysis group :


Gunther

Roland (MIT)

Christof

Roland (MIT)

Yen
-
Jie

Lee (MIT)

Mathews Nguyen (CERN)

Marguerite
Tonjes

(UMD),
Ying Lu (UMD)

Back up slides

13

CMS detector

Silicon Tracker


<2.4

D.
Dutta

, India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

14

EM and HAD Calorimeter

HF



<5

Beam
Scintillator

Counter

15


Calorimeter Jets

Particle
flow reconstructs all stable particle in the event: h
+/
-
, γ, h
0
, e,
μ

with thorough combination of all sub
-
detectors in CMS












On average jets are:

~ 65% charged hadrons, ~ 25% photons, ~ 10 % neutral hadrons

• Using the silicon tracker (vs. HCAL) to measure charged
hadrons


Particle Flow Jets

D.
Dutta

et al., India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

16

Inclusion Of Jet Triggers

Jet triggers are used to enhance the p
T

reach and to have low fake

Jet energy distribution

Charged particle distribution

D.
Dutta

et al., India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

17

A. Yoon, QM2011

D.
Dutta

et al., India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

Dijet Asymmetry

18

Striking enhancement of asymmetry
in data with increased centrality

CMS: arXiv:1102.1957

Dijet selection:



<2.0

leading jet
p
T
>100
GeV

Sub
-
leading jet
p
T
>40
GeV


>2*

/3


Dijet Fragmentation fn:

Studies on Dijet in CMS

Fragmentation pattern
independent of energy lost in
medium

)
/
(
ln
track
T
jet
T
p
p



d
dN
/
M.
Tonjes
, QM2011

19

Collinear
-

and Infrared
-
Safe



collinear splitting shouldn’t change jets



soft emissions shouldn’t change jets

Jet Algorithms

•Cone
-
Type Algorithms


Midpoint Cone (
Tev
), Iterative Cone (CMS),
SISCone

(LHC)



Typically not Infrared
-

& Collinear
-
Safe (exception:
SISCone
)



Typically complex,
invoving

several (non
-
phyiscal
) parameters



Favored at
hadron

colliders (computational performance?)



Strongly disfavored by theorists


•Sequential Clustering Algorithms



kT
, Cambridge/Aachen, Anti
-
kT



Infrared
-

& Collinear
-
Safe by construction



Clean & Simple Algorithms



Strongly favored by theorists



Not widely used at
hadron

colliders in the past



computational performance
(SOLVED)



jet area not trivially accessible
(SOLVED)

Collinear
-
Safety

Infrared
-
Safety


FastJet

implementation

C++ library providing
fast (!) JA

Implementation



kT
, Cambridge/Aachen, Anti
-
kT
,

SISCone



Sequential Clustering: yielding bit
identical results
w.r.t
. prior

implementations featuring

dramatically improved performance

20



Plan: the jet corrections will be factorized

-

Correcting for each factor in a fixed sequence up to a level chosen by the user.



Factorization facilitates the use of data
-
driven corrections

-

Breaking the correction into pieces that are naturally measured in collider data:


Offset: pile
-
up and noise measured in zero
-
bias events.


Relative: jet response vs. η relative to barrel found using
dijet

balance.


Absolute: jet response vs. PT found in barrel using γ / Z + jet.

-

Allows data
-
driven corrections as they emerge to easily replace MC truth



Pile
-
up

-
Refers to the energy from additional proton
-
proton collisions,

occurring close enough in time to the hard scatter to be

included in the calorimeter energy within the jet

-
Statistically independent: not correlated with hard scatter

-
Increases with luminosity

-
The additional energy amounts to 2.5
GeV
/10
GeV
/200
GeV

in a cone of radius 0.5 in the barrel for low luminosity pile
-
up,
high luminosity pile
-
up and heavy ion collisions, respectively.

Calorimeter noise

-
Refers to any noise above the
calorimeter cell and tower

thresholds for calorimeter
towers included in the jet


Both pile
-
up and electronic
noise produce an energy offset

Reconstructed
Jets

Offset

L1

Rel



L2

Abs
p
T

L3

Required Corrections

Optional Corrections

Jet Correction

EMF

Flavour

UE

Parton

Callibrated

jets

Pile
-
up subtracted

21

D.
Dutta

et al., India CMS Meet, BARC, Mumbai , 28
-
29 July 2011

21

M.
Naguyen
, QM2011

1
. Event
-
by
-
event subtraction of the

heavy
-
ion background


2. Jet energy corrections (JEC) based on

GEANT simulation of PYTHIA jets


3. Validation of the BG subtraction + JEC

for PYTHIA jets embedded in HYDJET

Jet reconstruction in heavy
ions

Background subtraction