Rick Field
December 1, 2004
HERWIG, JIMMY, and PYTHIA Tune A
Jon Butterworth
Jeff Forshaw
Mike Seymour
Outline of Talk
• Discuss the definition of the “underlying event” (UE).
• Use HERWIG to identify the particles coming from the “beam-beam
remnants” (BBR) and the particles coming from initial-state radiation
(ISR).
• Examine some differences between HERWIG and PYTHIA Tune A.
• Tune JIMMY multiple parton interactions (MPI) to agree with
PYTHIA Tune A.
• Study the BBR, MPI, and ISR contribution to the “transverse” region
and to “jets”.
• Discuss problems with JIMMY (i.e. no BBR, too “soft”, Q2
dependence).
TeV4LHC Meeting
Page 1 of 27
Rick Field
December 1, 2004
The “Underlying Event”
• What is your definition of the “underlying event” in a hard scattering
process?
Outgoing Parton
Acronym
Definition
2-to-2
Two outgoing partons
ISR
Initial State radiation
FSR
Final State Radiation
HARD
2-to-2 + ISR + FSR
BBR
Bean-Beam Remnants
MPI
Multiple Parton Interactions
Pile-up
Additional proton-antiproton collisions
MB
“Minimum-Bias” collisions
UE(1)
UE(2)
BBR + MPI
BBR + MPI + ISR
UE(3)
BBR + MPI + ISR + FSR
UE(4)
MB (does not make sense!)
HARD + UE(1)
PT(hard)
Initial-State Radiation
Proton
AntiProton
Underlying Event
Underlying Event
Outgoing Parton
Final-State
Radiation
Not equal to UE(1),
UE(2), or UE(3)!
“Min-Bias” Collisions
Hadron
Hadron
• My definition is UE(1), but for some jet corrections you might want
UE(2) or UE(3). No observable directly measures UE(1)!
• Drell-Yan and low pT Z-boson production measure UE(2)!
TeV4LHC Meeting
Page 2 of 27
Rick Field
December 1, 2004
The “Transverse” Region
Charged Particles
(pT > 0.5 GeV/c, |η| < 1)
Use the leading jet to define
the “transverse” region.
Look at the density of
charged particles in the
“transverse” region.
2π
Jet #1 Direction
Away Region
∆φ
Transverse
Region
“Toward”
“Transverse”
“Transverse”
φ
Leading
Jet
Toward Region
“Away”
Transverse
Region
HARD + UE(1)
contribute to the
“transverse” region!
Away Region
0
-1
η
+1
• Look at the “transverse” region as defined by the leading calorimeter jet
(JetClu R = 0.7, |η| < 2).
• Study the charged particles (pT > 0.5 GeV/c, |η| < 1) and form the charged
particle density, dNchg/dηdφ, and the charged scalar pT sum density,
dPTsum/dηdφ.
• Note that the “transverse” region is not UE(1) or UE(2) or UE(3)!
TeV4LHC Meeting
Page 3 of 27
Rick Field
December 1, 2004
“MAX/MIN Transverse” Densities
Jet #1 Direction
Jet #1 Direction
“Toward-Side” Jet
∆φ
∆φ
“Toward”
“Toward”
“TransMAX”
“TransMIN” very sensitive to
the “beam-beam remnants”!
“TransMIN”
“TransMAX”
“TransMIN”
Jet #3
“Away”
“Away”
“Away-Side” Jet
• Define the MAX and MIN “transverse” regions on an event-by-event basis with
MAX (MIN) having the largest (smallest) density.
• The “transMIN” region is very sensitive to the “beam-beam remnant” and
multiple parton interaction components of the “underlying event”.
• The difference, “transMAX” minus “transMIN”, is very sensitive to the “hard
scattering” component of the “underlying event” (i.e. hard initial and finalstate radiation).
TeV4LHC Meeting
Page 4 of 27
Rick Field
December 1, 2004
Leading Jet: “MAX & MIN Transverse” Densities
PYTHIA Tune A
HERWIG
"MAX/MIN Transverse" Charge Density: dN/dηdφ
"MAX/MIN Transverse" Charge Density: dN/dηdφ
1.6
CDF Preliminary
Leading Jet
"MAX"
data uncorrected
theory + CDFSIM
1.2
PYTHIA Tune A 1.96 TeV
"Transverse" Charge Density
"Transverse" Charge Density
1.6
"AVE"
0.8
"MIN"
0.4
CDF Preliminary
Leading Jet
1.2
"AVE"
0.8
"MIN"
0.4
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0.0
0
50
100
150
200
250
0
50
ET(jet#1) (GeV)
100
150
200
250
ET(jet#1) (GeV)
"MAX/MIN Transverse" PTsum Density: dPT/dηdφ
"MAX/MIN Transverse" PTsum Density: dPT/dηdφ
2.5
PYTHIA Tune A 1.96 TeV
CDF Preliminary
"MAX"
data uncorrected
theory + CDFSIM
2.0
Leading Jet
1.5
"AVE"
1.0
"MIN"
0.5
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
ET(jet#1) (GeV)
200
250
"Transverse" PTsum Density (GeV/c)
2.5
"Transverse" PTsum Density (GeV/c)
HERWIG 1.96 TeV
"MAX"
data uncorrected
theory + CDFSIM
HERWIG 1.96 TeV
CDF Preliminary
data uncorrected
theory + CDFSIM
2.0
"MAX"
Leading Jet
1.5
"AVE"
1.0
"MIN"
0.5
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
200
250
ET(jet#1) (GeV)
Charged particle density and PTsum density for “leading jet” events versus ET(jet#1) for PYTHIA Tune A and HERWIG.
TeV4LHC Meeting
Page 5 of 27
Rick Field
December 1, 2004
HERWIG agrees
with the data for
ET(jet#1) > 100
Leading Jet: “MAX & MIN Transverse” Densities
PYTHIA Tune A
HERWIG
"MAX/MIN Transverse" Charge Density: dN/dηdφ
"MAX/MIN Transverse" Charge Density: dN/dηdφ
1.6
CDF Preliminary
Leading Jet
"MAX"
data uncorrected
theory + CDFSIM
1.2
PYTHIA Tune A 1.96 TeV
"Transverse" Charge Density
"Transverse" Charge Density
1.6
"AVE"
0.8
"MIN"
0.4
CDF Preliminary
Leading Jet
1.2
"AVE"
0.8
"MIN"
0.4
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0.0
0
50
100
150
200
250
0
50
ET(jet#1) (GeV)
100
150
200
250
ET(jet#1) (GeV)
"MAX/MIN Transverse" PTsum Density: dPT/dηdφ
"MAX/MIN Transverse" PTsum Density: dPT/dηdφ
2.5
PYTHIA Tune A 1.96 TeV
CDF Preliminary
"MAX"
data uncorrected
theory + CDFSIM
2.0
Leading Jet
1.5
"AVE"
1.0
"MIN"
0.5
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
ET(jet#1) (GeV)
200
250
"Transverse" PTsum Density (GeV/c)
2.5
"Transverse" PTsum Density (GeV/c)
HERWIG 1.96 TeV
"MAX"
data uncorrected
theory + CDFSIM
HERWIG 1.96 TeV
CDF Preliminary
data uncorrected
theory + CDFSIM
2.0
"MAX"
Leading Jet
1.5
"AVE"
1.0
"MIN"
0.5
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
200
250
ET(jet#1) (GeV)
Charged particle density and PTsum density for “leading jet” events versus ET(jet#1) for PYTHIA Tune A and HERWIG.
TeV4LHC Meeting
Page 6 of 27
Rick Field
December 1, 2004
HERWIG has too few
“Leading Jet” Charge Density: 30 < ET(jet#1) < 90charged
GeVparticles in the
PYTHIA Tune A
Charged Particle Density: dN/dηdφ
CDF Preliminary
Leading Jet
data uncorrected
theory + CDFSIM
PY Tune A
Charged Particle Density: dN/dηdφ
10.0
30 < ET(jet#1) < 70 GeV
CDF Preliminary
Charged Particle Density
Charged Particle Density
10.0
1.0
"Transverse"
Region
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
Jet#1
0.1
data uncorrected
theory + CDFSIM
30 < ET(jet#1) < 70 GeV
Leading Jet
HERWIG
1.0
"Transverse"
Region
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
Jet#1
0.1
0
30
60
90
120
150
180
210
240
270
300
330
360
0
30
60
90
120
∆φ (degrees)
data uncorrected
theory + CDFSIM
0.5
Leading Jet
30 < ET(jet#1) < 70 GeV
CDF Preliminary
PYTHIA Tune A
0.0
"Transverse"
Region
0
30
60
90
data uncorrected
theory + CDFSIM
180
210
∆φ (degrees)
300
330
360
240
270
300
330
360
HERWIG
"Transverse"
Region
Jet#1
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
-1.0
150
270
0.0
Jet#1
120
240
Leading Jet
30 < ET(jet#1) < 70 GeV
-0.5
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
-1.0
210
Data - Theory: Charged Particle Density dN/dηdφ
0.5
-0.5
180
1.0
Data - Theory
CDF Preliminary
150
∆φ (degrees)
Data - Theory: Charged Particle Density dN/dηdφ
1.0
Data - Theory
“transverse” region!
HERWIG
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
Charged particle density for “leading jet” events with 30 < ET(jet#1) < 70 GeV versus PYTHIA Tune A and HERWIG.
TeV4LHC Meeting
Page 7 of 27
Rick Field
December 1, 2004
HERWIG does not have
enough charged PTsum in
the “transverse” region!
PYTHIA Tune A vs HERWIG: PTsum Density dPT/dηdφ
PYTHIA Tune A
HERWIG
Charged PTsum Density: dPT/dηdφ
Charged PTsum Density: dPT/dηdφ
100.0
Charged Particles
30 < ET(jet#1) < 70 GeV
(|η|<1.0, PT>0.5 GeV/c)
Leading Jet
PY Tune A
Charged PTsum Density (GeV/c)
Charged PTsum Density (GeV/c)
100.0
10.0
1.0
CDF Preliminary
0.1
0
30
60
Jet#1
"Transverse"
Region
data uncorrected
theory + CDFSIM
HERWIG
10.0
"Transverse"
Region
1.0
CDF Preliminary
120
150
180
210
240
270
300
330
360
Jet#1
data uncorrected
theory + CDFSIM
0.1
90
Charged Particles
30 < ET(jet#1) < 70 GeV
(|η|<1.0, PT>0.5 GeV/c)
Leading Jet
0
30
60
90
120
∆φ (degrees)
Leading Jet
30 < ET(jet#1) < 70 GeV
data uncorrected
theory + CDFSIM
1
240
270
-1
"Transverse"
Region
Leading Jet
30 < ET(jet#1) < 70 GeV
CDF Preliminary
PYTHIA Tune A
0
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
210
300
330
360
Data - Theory: Charged PTsum Density dPT/dηdφ
2
Data - Theory (GeV/c)
Data - Theory (GeV/c)
CDF Preliminary
180
∆φ (degrees)
Data - Theory: Charged PTsum Density dPT/dηdφ
2
150
data uncorrected
theory + CDFSIM
1
0
-1
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
Jet#1
-2
HERWIG
Jet#1
"Transverse"
Region
-2
0
30
60
90
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
Charged PTsum density for “leading jet” events with 30 < ET(jet#1) < 70 GeV versus PYTHIA Tune A and HERWIG.
TeV4LHC Meeting
Page 8 of 27
Rick Field
December 1, 2004
HERWIG: PTsum Density dPT/dηdφ
Charged PTsum Density: dPT/dηdφ
RDF Preliminary
PTsum Density (GeV/c)
generator level
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
PTsum Density: dPT/dηdφ
100.0
PT(jet#1) > 30 GeV/c
PTsum Density (GeV/c)
100.0
HW TOT
HW 2-to-2
HW ISR
HW BBR
10.0
"Transverse"
Region
1.0
HW TOT
HW 2-to-2
HW ISR
HW BBR
10.0
1.0
generator level
0.1
60
90
120
150
180
210
∆φ (degrees)
240
270
Jet#1
RDF Preliminary
0.1
30
PT(jet#1) > 30 GeV/c
"Transverse"
Region
Jet#1
0
All Particles
(|η|<1.0, PT>0 GeV/c)
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of HERWIG for the ∆φ dependence
of the charged scalar PTsum density (|η|<1, pT>0.5 GeV/c) relative to the
leading jet for PT(jet#1) > 30 GeV/c.
• (right) Shows the generator level predictions of HERWIG for the ∆φ
dependence of the overall scalar PTsum density (|η|<1, pT>0) relative to the
leading jet for PT(jet#1) > 30 GeV/c.
• The contributions from the “beam-beam remnants” (BBR), initial-state
radiation (ISR), and the 2-to-2 hard scattering plus finial-state radiation (2-to2+FSR) are shown.
TeV4LHC Meeting
Page 9 of 27
Rick Field
December 1, 2004
Leading jet is trying to
“suck up” the BBR
without much success!
HERWIG: PTsum Density dPT/dηdφ
PTsum Density: dPT/dηdφ
PTsum Density: dPT/dηdφ
100.0
1.0
generator level
RDF Preliminary
HERWIG All Particles
(|η|<1.0, PT>0 GeV/c)
10.0
"Transverse"
Region
1.0
PTsum Density (GeV/c)
PTsum Density (GeV/c)
RDF Preliminary
Total
BBR
Jet#1
solid = PT(jet#1) > 30 GeV/c
open = PT(jet#1) > 60 GeV/c
0.1
0
30
60
90
120
generator level
All Particles
(|η|<1.0, PT>0 GeV/c)
HERWIG
Beam-Beam Remnants
0.8
0.6
0.4
solid = PT(jet#1) > 30 GeV/c
open = PT(jet#1) > 60 GeV/c
ISR
"Transverse"
Region
Jet#1
0.2
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of HERWIG for the ∆φ dependence
of the overall scalar PTsum density (|η|<1, pT>0) relative to the leading jet for
PT(jet#1) > 30 GeV/c and for PT(jet#1) > 60 GeV/c.
• (right) Shows the generator level predictions of HERWIG for the ∆φ
dependence of the “beam-beam remnant” (BBR) contribution to the overall
PTsum density (|η|<1, pT>0) relative to the leading jet for PT(jet#1) > 30 GeV/c
and for PT(jet#1) > 60 GeV/c. I was wrong! For HERWIG the BBR component
is, on the average, essentially azimuthally symmetric (not enough pT for the
leading jet to “suck up”).
TeV4LHC Meeting
Page 10 of 27
Rick Field
December 1, 2004
100 MeV
difference for
BBR!
HERWIG: ETsum Density dET/dηdφ
ETsum Density: dET/dηdφ
HW TOT
HW 2-to-2
HW ISR
HW BBR
10.0
Difference: dET/dηdφ - dPT/dηdφ
0.4
All Particles
(|η|<1.0, PT>0 GeV/c)
RDF Preliminary
PT(jet#1) > 30 GeV
Difference: ET - cPT (GeV)
ETsum Density (GeV)
100.0
"Transverse"
Region
1.0
Jet#1
RDF Preliminary
generator level
0.3
HW TOT
PT(jet#1) > 30 GeV/c
HW 2-to-2
HW ISR
HW BBR
All Particles
(|η|<1.0, PT>0 GeV/c)
"Transverse"
Region
0.2
0.1
Jet#
generator level
0.1
0.0
0
30
60
90
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of HERWIG for the ∆φ dependence
of the overall scalar ETsum density (|η|<1, pT>0) relative to the leading jet for
ET(jet#1) > 30 GeV/c.
• (right) Shows the generator level predictions of HERWIG for the ∆φ
dependence of the ETsum density minus the PTsum density (|η|<1, pT>0)
relative to the leading jet for PT(jet#1) > 30 GeV/c.
• The contributions from the “beam-beam remnants” (BBR), initial-state
radiation (ISR), and the 2-to-2 hard scattering plus finial-state radiation (2-to2+FSR) are shown.
TeV4LHC Meeting
Page 11 of 27
Rick Field
December 1, 2004
PYTHIA Tune A
and HERWIG are
different!
PYTHIA Tune A vs HERWIG: PTsum Density dPT/dηdφ
Density Ratio: All(pT>0)/Charged(pT>0.5 GeV/c)
PTsum Density: dPT/dηdφ
PYA ALL (PT>0)
HW ALL (PT>0)
PYA CHG (PT>0.5 GeV/c)
HW CHG (PT>0.5 GeV/c)
10.0
6
PT(jet#1) > 30 GeV/c
Particles
(|η|<1.0)
"Transverse"
Region
1.0
RDF Preliminary
4
3
2
generator level
0
0
30
60
90
120
150
180
210
∆φ (degrees)
240
270
300
330
360
Jet#1
RDF Preliminary
generator level
0.1
PT(jet#1) > 30 GeV/c
"Transverse"
Region
1
Jet#1
Particles
(|η|<1.0)
PYA TOT
HW TOT
HW BBR
5
PTsum Density Ratio
PTsum Density (GeV/c)
100.0
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and HERWIG
for the ∆φ dependence of the charged scalar PTsum density (|η|<1, pT>0.5
GeV/c) and the overall scalar PTsum density (|η|<1, pT>0) relative to the
leading jet for PT(jet#1) > 30 GeV/c.
• (right) Ratio of the overall PTsum density (|η|<1, pT>0) to the charged PTsum
density (|η|<1, pT>0.5 GeV/c) relative to the leading jet for PYTHIA Tune A
and HERWIG. The BBR component of HERWIG is also shown.
• Note that PYTHIA Tune A and HERWIG differ in the extrapolation from pT >
0.5 GeV/c to pT > 0!
TeV4LHC Meeting
Page 12 of 27
Rick Field
December 1, 2004
HERWIG: Number Density dN/dηdφ
Number Density: dN/dηdφ
Charged Particle Density: dN/dηdφ
10.0
100.0
HW TOT
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
1.0
90
120
1.0
RDF Preliminary
Jet#
"Transverse"
Region
generator level
generator level
60
HW BBR
10.0
Jet#1
RDF Preliminary
30
PT(jet#1) > 30 GeV/c
HW ISR
"Transverse"
Region
0
All Particles
(|η|<1.0, PT>0 GeV/c)
HW 2-to-2
HW ISR
HW BBR
0.1
HW TOT
PT(jet#1) > 30 GeV
Number Density
Charged Particler Density
HW 2-to-2
0.1
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of HERWIG for the ∆φ dependence
of the charged particle density (|η|<1, pT>0.5 GeV/c) relative to the leading jet
for PT(jet#1) > 30 GeV/c.
• (right) Shows the generator level predictions of HERWIG for the ∆φ
dependence of the overall particle density (|η|<1, pT>0) relative to the leading
jet for PT(jet#1) > 30 GeV/c.
• The contributions from the “beam-beam remnants” (BBR), initial-state
radiation (ISR), and the 2-to-2 hard scattering plus finial-state radiation (2-to2+FSR) are shown.
TeV4LHC Meeting
Page 13 of 27
Rick Field
December 1, 2004
PYTHIA Tune A
PYTHIA Tune A vs HERWIG: Number Density dN/dηdφ
and HERWIG are
different!
Density Ratio: All(pT>0)/Charged(pT>0.5 GeV/c)
Particle Density: dN/dηdφ
100.0
10
PYA ALL (PT>0)
Particles
(|η|<1.0)
HW ALL (PT>0)
PT(jet#1) > 30 GeV/c
10.0
Number Density Ratio
Number Density
PYA CHG (PT>0.5 GeV/c)
HW CHG (PT>0.5 GeV/c)
"Transverse"
Region
1.0
RDF Preliminary
0
30
60
90
HW TOT
8
PT(jet#1) > 30 GeV/c
HW BBR
6
4
2
RDF Preliminary
Jet#1
"Transverse"
Region
generator level
generator level
0.1
Particles
(|η|<1.0)
PYA TOT
Jet#
0
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and HERWIG
for the ∆φ dependence of the charged particle density (|η|<1, pT>0.5 GeV/c) and
the overall particle density (|η|<1, pT>0) relative to the leading jet for PT(jet#1)
> 30 GeV/c.
• (right) Ratio of the overall particle density (|η|<1, pT>0) to the charged particle
density (|η|<1, pT>0.5 GeV/c) relative to the leading jet for PYTHIA Tune A
and HERWIG. The BBR component of HERWIG is also shown.
• Note that PYTHIA Tune A and HERWIG differ in the extrapolation from pT >
0.5 GeV/c to pT > 0!
TeV4LHC Meeting
Page 14 of 27
Rick Field
December 1, 2004
HERWIG BBR
adjusted to agree
with PYTHIA Tune
PYTHIA Tune A vs HERWIG: “Transverse Region”
Charged PTsum Density: dPT/dηdφ
RDF Preliminary
generator level
PYA TOT
Charged PTsum Density: dPT/dηdφ
100.0
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
PT(jet#1) > 30 GeV/c
Charged PTsum Density (GeV/c)
Charged PTsum Density (GeV/c)
100.0
HW TOT
10.0
HW 2-to-2
"Transverse"
Region
HW ISR
HW BBR
1.0
Jet#1
0.1
RDF Preliminary
generator level
PYA TOT
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
PT(jet#1) > 30 GeV/c
HW TOT
10.0
HW 2-to-2
"Transverse"
Region
HW ISR
HW BBR
1.0
BBR x 2
Jet#1
0.1
0
30
60
90
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and HERWIG
for the ∆φ dependence of the charged scalar PTsum density (|η|<1, pT>0.5
GeV/c) relative to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Same as (left) but with HERWIG BBR increased by a factor of 2.0. The
multiple parton interactions in PYTHIA Tune A result in a factor of 2 increase
in the PTsum of BBR charged particles (pT>0.5 GeV/c) over HERWIG!
• HW “Low” = BBR plus HERWIG extrapolation to pT = 0.
• HW “High” = BBR x 2.0 plus HERWIG extrapolation to pT = 0.
• HW “High” agrees with PYTHIA Tune A for the charged PTsum (pT>0.5
GeV/c) in the “transverse” region!
TeV4LHC Meeting
Page 15 of 27
Rick Field
December 1, 2004
HERWIG BBR
adjusted to agree
with PYTHIA Tune
PYTHIA Tune A vs HERWIG: “Transverse Region”
Charged Particle Density: dN/dηdφ
PYA TOT
HW TOT
HW 2-to-2
HW ISR
HW BBR
Charged Particle Density: dN/dηdφ
10.0
PT(jet#1) > 30 GeV
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
Number Density
Number Density
10.0
"Transverse"
Region
1.0
PYA TOT
HW TOT
HW 2-to-2
HW ISR
HW BBR
PT(jet#1) > 30 GeV
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
"Transverse"
Region
1.0
BBR x 1.4
RDF Preliminary
RDF Preliminary
Jet#1
generator level
0.1
0
30
60
90
120
0.1
150
180
210
∆φ (degrees)
240
270
300
330
360
Jet#1
generator level
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and HERWIG
for the ∆φ dependence of the charged particle density (|η|<1, pT>0.5 GeV/c)
relative to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Same as (left) but with HERWIG BBR increased by a factor of 1.4.
• HERWIG now agrees with PYTHIA Tune A in the “transverse” region. The
multiple parton interactions in PYTHIA Tune A result in a 40% increase in the
number of BBR charged particles (pT>0.5 GeV/c) over HERWIG!
• HERWIG particles are “softer” than PYTHIA Tune A! Cannot make both the
charged particle density (pT>0.5 GeV) and the charged PTsum density agree
with PYTHIA Tune A!
TeV4LHC Meeting
Page 16 of 27
Rick Field
December 1, 2004
HERWIG BBR
adjusted to agree
with PYTHIA Tune
PYTHIA Tune A vs HERWIG: “Transverse Region”
PTsum Density: dPT/dηdφ
PYA TOT
HW TOT
HW 2-to-2
All Particles
(|η|<1.0, PT>0 GeV/c)
HW ISR
HW BBR
10.0
PTsum Density: dPT/dηdφ
100.0
PT(jet#1) > 30 GeV/c
PTsum Density (GeV/c)
PTsum Density (GeV/c)
100.0
"Transverse"
Region
1.0
30
60
90
150
180
210
∆φ (degrees)
240
270
"Transverse"
Region
1.0
BBR x 1.4
Jet#1
generator level
0.1
120
PT(jet#1) > 30 GeV/c
RDF Preliminary
generator level
0
All Particles
(|η|<1.0, PT>0 GeV/c)
HW ISR
HW BBR
10.0
Jet#1
RDF Preliminary
0.1
PYA TOT
HW TOT
HW 2-to-2
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and HERWIG
for the ∆φ dependence of the overall scalar PTsum density (|η|<1, pT>0) relative
to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Same as (left) but with HERWIG BBR increased by a factor of 1.4. The
multiple parton interactions in PYTHIA Tune A result in a 40% increase in the
PTsum of all BBR particles (pT>0) over HERWIG!
• HW “Mid” = BBR x 1.4 (assumes PYTHIA Tune A extrapolation to pT = 0).
• HW “Mid” agrees with PYTHIA Tune A for the overall PTsum (pT>0) in the
“transverse” region!
TeV4LHC Meeting
Page 17 of 27
Rick Field
December 1, 2004
JIMMY tuned to
agree with
PYTHIA Tune A!
PYTHIA Tune A vs JIMMY: “Transverse Region”
Charged PTsum Density: dPT/dηdφ
Charged PTsum Density: dPT/dηdφ
100.0
Charged Particles
30 < ET(jet#1) < 70 GeV
(|η|<1.0, PT>0.5 GeV/c)
Leading Jet
PY Tune A
Charged PTsum Density (GeV/c)
Charged PTsum Density (GeV/c)
100.0
10.0
1.0
CDF Preliminary
0.1
0
30
60
90
Jet#1
"Transverse"
Region
data uncorrected
theory + CDFSIM
RDF Preliminary
generator level
PYA TOT
JM TOT
10.0
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
JM 2-to-2
JM ISR
JM MPI
PT(jet#1) > 30 GeV/c
"Transverse"
Region
1.0
Jet#1
0.1
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the Run 2 data on the ∆φ dependence of the charged scalar PTsum
density (|η|<1, pT>0.5 GeV/c) relative to the leading jet for 30 < ET(jet#1) < 70
GeV/c compared with PYTHIA Tune A (after CDFSIM).
• (right) Shows the generator level predictions of PYTHIA Tune A and JIMMY
(PTmin=1.8 GeV/c) for the ∆φ dependence of the charged scalar PTsum density
(|η|<1, pT>0.5 GeV/c) relative to the leading jet for PT(jet#1) > 30 GeV/c.
JIMMY and PYTHIA Tune A agree in the “transverse” region.
• For JIMMY the contributions from the multiple parton interactions (MPI),
initial-state radiation (ISR), and the 2-to-2 hard scattering plus finial-state
radiation (2-to-2+FSR) are shown.
TeV4LHC Meeting
Page 18 of 27
Rick Field
December 1, 2004
Leading Jet “sucksup” MPI particles!
JIMMY (MPI) vs HERWIG (BBR)
JIMMY: Number of MPI Scatterings
Charged PTsum Density: dPT/dηdφ
0.14
PT(jet#1) > 30 GeV/c
0.8
generator level
0.6
JM MPI
HW BBR
"Transverse"
Region
0.4
0.2
0.0
0
30
60
90
generator level
0.10
PT(jet#1) > 30 GeV/c
0.08
0.06
0.04
0.02
Jet#1
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
RDF Preliminary
0.12
Fraction of Events
Charged PTsum Density (GeV/c)
1.0
RDF Preliminary
No BBR and no MPI!
0.00
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20
Number of MPI Scatterings
• (left) Shows the generator level predictions of JIMMY (MPI, PTmin=1.8 GeV/c)
and HERWIG (BBR) for the ∆φ dependence of the charged scalar PTsum
density (|η|<1, pT>0.5 GeV/c) relative to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Shows the predictions of JIMMY (MPI, PTmin=1.8 GeV/c) for the
number of multiple parton interactions (MPI) for PT(jet#1) > 30 GeV/c.
• Note that JIMMY has no BBR. When the number of MPI is equal to zero the
event has no BBR and no MPI! This cannot be right! JIMMY should include
HERWIG BBR whenever the number of MPI scatterings is zero.
• I tried to make HERWIG produce BBR whenever JIMMY produced no MPI
scattering but I was not successful (need help from the authors!).
TeV4LHC Meeting
Page 19 of 27
Rick Field
December 1, 2004
JIMMY (MPI) vs HERWIG (BBR)
Charged PTsum Density: dPT/dηdφ
ETsum Density: dET/dηdφ
2.5
0.8
generator level
0.6
JM MPI
HW BBR
"Transverse"
Region
0.4
0.2
0
30
60
90
generator level
2.0
150
180
210
∆φ (degrees)
240
270
300
330
360
PT(jet#1) > 30 GeV
1.0
0.5
0
30
60
90
Jet#1
"Transverse"
Region
All Particles
(|η|<1.0, PT>0 GeV/c)
0.0
120
JM MPI
HW BBR
1.5
Jet#1
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
0.0
RDF Preliminary
PT(jet#1) > 30 GeV/c
RDF Preliminary
ETsum Density (GeV)
Charged PTsum Density (GeV/c)
1.0
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of JIMMY (MPI, PTmin=1.8 GeV/c)
and HERWIG (BBR) for the ∆φ dependence of the charged scalar PTsum
density (|η|<1, pT>0.5 GeV/c) relative to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Shows the generator level predictions of JIMMY (MPI, PTmin=1.8
GeV/c) and HERWIG (BBR) for the ∆φ dependence of the scalar ETsum
density (|η|<1, pT>0 GeV/c) relative to the leading jet for PT(jet#1) > 30 GeV/c.
• For PTsum in the “transverse” region JIMMY (MPI) is similar to HW “High”
(i.e. HERWIG BBR x 2.0)!
TeV4LHC Meeting
Page 20 of 27
Rick Field
December 1, 2004
Tuned JIMMY produces
more charged particles
than PYTHIA Tune A!
JIMMY tuned to
agree with
PYTHIA Tune A!
PYTHIA Tune A vs JIMMY: “Transverse Region”
Charged PTsum Density: dPT/dηdφ
RDF Preliminary
generator level
PYA TOT
JM TOT
10.0
Charged Particle Density: dN/dηdφ
10.0
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
JM 2-to-2
JM ISR
JM MPI
PYA TOT
JM TOT
JM 2-to-2
JM ISR
JM MPI
PT(jet#1) > 30 GeV/c
Number Density
Charged PTsum Density (GeV/c)
100.0
"Transverse"
Region
1.0
RDF Preliminary
0.1
60
90
120
150
180
210
∆φ (degrees)
240
270
300
330
360
Jet#1
generator level
0.1
30
"Transverse"
Region
1.0
Jet#1
0
PT(jet#1) > 30 GeV
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
0
30
60
90
120
150
180
210
240
270
300
330
360
∆φ (degrees)
• (left) Shows the generator level predictions of PYTHIA Tune A and JIMMY
(PTmin=1.8 GeV/c) for the ∆φ dependence of the charged scalar PTsum density
(|η|<1, pT>0.5 GeV/c) relative to the leading jet with PT(jet#1) > 30 GeV/c.
JIMMY and PYTHIA Tune A agree in the “transverse” region.
• (right) Shows the generator level predictions of PYTHIA Tune A and JIMMY
(PTmin=1.8 GeV/c) for the ∆φ dependence of the charged particle density (|η|<1,
pT>0.5 GeV/c) relative to the leading jet with PT(jet#1) > 30 GeV/c.
• By only varying PTmin I cannot make JIMMY and PYTHIA Tune A agree for
both the PTsum density and the number density. JIMMY produces a softer pT
distribution.
TeV4LHC Meeting
Page 21 of 27
Rick Field
December 1, 2004
Tuned JIMMY
produces more ETsum
than PYTHIA Tune A!
PYTHIA Tune A vs JIMMY: “Transverse Region”
ETsum Density: dET/dηdφ
PYA TOT
JM TOT
JM 2-to-2
JM ISR
10.0
All Particles
(|η|<1.0, PT>0 GeV/c)
JM MPI
"Transverse" PTsum Density (GeV/c)
ETsum Density (GeV)
"Transverse" PTsum Density: dPT/dηdφ
4
100.0
PT(jet#1) > 30 GeV
"Transverse"
Region
1.0
Jet#1
RDF Preliminary
generator level
0.1
0
30
60
90
RDF Preliminary
Jimmy
HW High
generator level
3
HW Mid
PYA = dashed
HW = solid
HW Low
PYA
2
1
All Particles
(|η|<1.0, PT>0 GeV/c)
1.96 TeV
0
120
150
180
210
∆φ (degrees)
240
270
300
330
360
0
50
100
150
200
250
300
350
400
450
500
PT(jet#1) (GeV/c)
• (left) Shows the generator level predictions of PYTHIA Tune A and JIMMY
(PTmin=1.8 GeV/c) for the ∆φ dependence of the scalar ETsum density (|η|<1,
pT>0) relative to the leading jet for PT(jet#1) > 30 GeV/c.
• (right) Shows the generator level predictions of PYTHIA Tune A (dashed),
HERWIG (high, mid, low), and JIMMY (PTmin=1.8 GeV/c) for overall scalar
PTsum density (|η|<1, pT>0) in the “transverse” region versus PT(jet#1).
• The tuned JIMMY is very similar to HERWIG “high” (i.e. BBR x 2.0).
• The tuned JIMMY produces a lot more ETsum (pT>0) in the “transverse”
region than does PYTHIA Tune A!
TeV4LHC Meeting
Page 22 of 27
Rick Field
December 1, 2004
PYTHIA Tune A vs JIMMY: “Transverse Region”
"Transverse" PTsum Density: dPT/dηdφ
"Transverse" PTsum Density: dPT/dηdφ
1.6
"Transverse" PTsum Density
RDF Preliminary
2.5
Max Transverse
generator level
PYA = dashed
JM = solid
2.0
1.96 TeV
Average Transverse
1.5
1.0
Min Transverse
0.5
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
0.0
"Transverse" PTsum Density (GeV/c)
3.0
Total
RDF Preliminary
generator level
1.2
PYA = dashed
JM = solid
1.96 TeV
0.8
ISR
2-to-2 + FSR
0.4
MPI
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
200
250
300
PT(jet#1) (GeV/c)
350
400
450
500
0
50
100
150
200
250
300
350
400
450
500
PT(jet#1) (GeV/c)
• (left) Shows the generator level predictions of PYTHIA Tune A (dashed) and
JIMMY (PTmin=1.8 GeV/c) for charged scalar PTsum density (|η|<1, pT>0.5
GeV/c) in the MAX/MIN/AVE “transverse” region versus PT(jet#1).
• (right) Shows the generator level predictions of PYTHIA Tune A (dashed) and
JIMMY (PTmin=1.8 GeV/c) for charged scalar PTsum density (|η|<1, pT>0.5
GeV/c) in the AVE “transverse” region versus PT(jet#1). For JIMMY the
contributions from the multiple parton interactions (MPI), initial-state
radiation (ISR), and the 2-to-2 hard scattering plus finial-state radiation (2-to2+FSR) are shown.
TeV4LHC Meeting
Page 23 of 27
Rick Field
December 1, 2004
PYTHIA Tune A vs JIMMY: “Transverse Region”
"Transverse" PTsum Density: dPT/dηdφ
"MAX/MIN Transverse" PTsum Density: dPT/dηdφ
3.0
PYTHIA Tune A 1.96 TeV
CDF Preliminary
2.0
RDF Preliminary
"MAX"
data uncorrected
theory + CDFSIM
"Transverse" PTsum Density
"Transverse" PTsum Density (GeV/c)
2.5
Leading Jet
1.5
"AVE"
1.0
"MIN"
0.5
Charged Particles (|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
ET(jet#1) (GeV)
200
250
2.5
Max Transverse
generator level
PYA = dashed
JM = solid
2.0
1.96 TeV
Average Transverse
1.5
1.0
Min Transverse
0.5
Charged Particles
(|η|<1.0, PT>0.5 GeV/c)
0.0
0
50
100
150
200
250
300
350
400
450
500
PT(jet#1) (GeV/c)
• (left) Run 2 data for charged scalar PTsum density (|η|<1, pT>0.5 GeV/c) in the
MAX/MIN/AVE “transverse” region versus PT(jet#1) compared with PYTHIA
Tune A (after CDFSIM).
• (right) Shows the generator level predictions of PYTHIA Tune A (dashed) and
JIMMY (PTmin=1.8 GeV/c) for charged scalar PTsum density (|η|<1, pT>0.5
GeV/c) in the MAX/MIN/AVE “transverse” region versus PT(jet#1).
• The tuned JIMMY now agrees with PYTHIA for PT(jet#1) < 100 GeV but
produces much more activity than PYTHIA Tune A (and the data?) in the
“transverse” region for PT(jet#1) > 100 GeV!
TeV4LHC Meeting
Page 24 of 27
Rick Field
December 1, 2004
BBR and MPI Contribution to Jets
JIMMY MPI!
Leading (R = 0.7) Jet: ETsum
3.0
Leading Jet ETsum (GeV)
RDF Preliminary
1.96 TeV |η(jet#1)|<2
generator level
2.5
JM MPI
2.0
HW BBR x 2
HW BBR x 1.4
1.5
1.0
HW BBR
HERWIG BBR
JIMMY MPI
0.5
0
50
100
150
200
250
300
All Particles
(PT>0 GeV/c)
350
400
450
500
PT(jet#1) (GeV/c)
• Shows the generator level predictions of HERWIG for the overall scalar
ETsum of BBR particles within jet#1 (leading jet) (pT>0 GeV/c) versus
pT(jet#1) for three levels of HERWIG BBR (high, mid, low).
• Shows the generator level predictions of JIMMY for the overall scalar ETsum
of MPI particles within jet#1 (leading jet) (pT>0 GeV/c) versus pT(jet#1) .
• Note that the BBR and MPI contribution to jets varies rapidly over the range
PT(jet) < 100 GeV/c.
TeV4LHC Meeting
Page 25 of 27
Rick Field
December 1, 2004
BBR and MPI Contribution to Jets
CDF Run 1 and Run 2
“jet UE” contribution
1.56 GeV ± 30%.
Leading (R = 0.7) Jet: ETsum
3.0
Leading Jet ETsum (GeV)
RDF Preliminary
1.96 TeV |η(jet#1)|<2
generator level
2.5
JM MPI
2.0
HW BBR x 2
Run I UE
HW BBR x 1.4
1.5
1.0
HW BBR
HERWIG BBR
JIMMY MPI
0.5
0
50
100
150
200
250
300
1.56 GeV
All Particles
(PT>0 GeV/c)
350
400
450
500
PT(jet#1) (GeV/c)
• Shows the generator level predictions of HERWIG for the overall scalar
ETsum of BBR particles within jet#1 (leading jet) (pT>0 GeV/c) versus
pT(jet#1) for three levels of HERWIG BBR (high, mid, low).
• Shows the generator level predictions of JIMMY for the overall scalar ETsum
of MPI particles within jet#1 (leading jet) (pT>0 GeV/c) versus pT(jet#1) .
• Note that the BBR and MPI contribution to jets varies rapidly over the range
PT(jet) < 100 GeV/c.
TeV4LHC Meeting
Page 26 of 27
Rick Field
December 1, 2004
Summary & Plans
• Problems with JIMMY:
ƒ No BBR or MPI when the number of MPI scatterings is equal to zero.
ƒ Too “soft”! Need to decrease the number of MPI scatterings and make
then “harder”.
ƒ Problem with the Q2 dependence. If you make things agree for 15 <
ET(jet#1) < 100 GeV, then too much UE activity for ET(jet#1) > 100 GeV!
• Next Step:
ƒ Work with the authors to fix the problems with JIMMY.
ƒ Study the calorimeter tower energy density in the “transverse region” and
compare with HERWIG, JIMMY, and PYTHIA Tune A.
ƒ Study Drell-Yan and low pT Z-boson production which looks directly at
UE(2) = BBR+MPI+ISR.
TeV4LHC Meeting
Page 27 of 27