CDF Collaboration Meeting

1
CDF Collaboration Meeting
Toward an Understanding of Hadron-Hadron
Collisions
Rick Field University of Florida
La Biodola, Elba Island, Tuscany, Italy
CDF Run 2
2
Toward and Understanding of Hadron-Hadron
Collisions
1st hat!
Feynman
and
Field

• From 7 GeV/c p0s to 600 GeV/c Jets.

• The Underlying Event at the Tevatron (things we
  dont understand).

• New Run 2 Monte-Carlo Tunes (extrapolations to
  the LHC).

• Lets find the Higgs!

3
The Feynman-Field Days
1973-1983
Feynman-Field Jet Model

• FF1 Quark Elastic Scattering as a Source of
  High Transverse Momentum Mesons, R. D. Field
  and R. P. Feynman, Phys. Rev. D15, 2590-2616
  (1977).
• FFF1 Correlations Among Particles and Jets
  Produced with Large Transverse Momenta, R. P.
  Feynman, R. D. Field and G. C. Fox, Nucl. Phys.
  B128, 1-65 (1977).
• FF2 A Parameterization of the properties of
  Quark Jets, R. D. Field and R. P. Feynman,
  Nucl. Phys. B136, 1-76 (1978).
• F1 Can Existing High Transverse Momentum Hadron
  Experiments be Interpreted by Contemporary
  Quantum Chromodynamics Ideas?, R. D. Field,
  Phys. Rev. Letters 40, 997-1000 (1978).
• FFF2 A Quantum Chromodynamic Approach for the
  Large Transverse Momentum Production of Particles
  and Jets, R. P. Feynman, R. D. Field and G. C.
  Fox, Phys. Rev. D18, 3320-3343 (1978).

• FW1 A QCD Model for ee- Annihilation, R. D.
  Field and S. Wolfram, Nucl. Phys. B213, 65-84
  (1983).

My 1st graduate student!
4
Before Feynman-Field
5
Before Feynman-Field
Rick Jimmie 1970
Rick Jimmie 1968
Rick Jimmie 1972 (pregnant!)
Rick Jimmie at CALTECH 1973
6
The Feynman-Field Days
Rick Field
Chris Quigg
Giorgio Bellettini
Keith Ellis
Keith Ellis
7
Hadron-Hadron Collisions
FF1 1977 (preQCD)

• What happens when two hadrons collide at high
  energy?

Feynman quote from FF1 The model we shall choose
is not a popular one, so that we will not
duplicate too much of the work of others who are
similarly analyzing various models (e.g.
constituent interchange model, multiperipheral
models, etc.). We shall assume that the high PT
particles arise from direct hard collisions
between constituent quarks in the incoming
particles, which fragment or cascade down into
several hadrons.

• Most of the time the hadrons ooze through each
  other and fall apart (i.e. no hard scattering).
  The outgoing particles continue in roughly the
  same direction as initial proton and antiproton.

• Occasionally there will be a large transverse
  momentum meson. Question Where did it come from?

• We assumed it came from quark-quark elastic
  scattering, but we did not know how to calculate
  it!

Black-Box Model
8
Quark-Quark Black-Box Model
No gluons!
FF1 1977 (preQCD)
Quark Distribution Functions determined from
deep-inelastic lepton-hadron collisions
Feynman quote from FF1 Because of the incomplete
knowledge of our functions some things can be
predicted with more certainty than others.
Those experimental results that are not well
predicted can be used up to determine these
functions in greater detail to permit better
predictions of further experiments. Our papers
will be a bit long because we wish to discuss
this interplay in detail.
Quark Fragmentation Functions determined from
ee- annihilations
Quark-Quark Cross-Section Unknown! Deteremined
from hadron-hadron collisions.
9
Quark-Quark Black-Box Model
Predict particle ratios
FF1 1977 (preQCD)
Predict increase with increasing CM energy W
Beam-Beam Remnants
Predict overall event topology (FFF1 paper 1977)
7 GeV/c p0s!
10
Telagram from Feynman
July 1976
SAW CRONIN AM NOW CONVINCED WERE RIGHT TRACK
QUICK WRITE FEYNMAN
11
Feynman Talk at Coral Gables (December 1976)
1st transparency
Last transparency
Feynman-Field Jet Model
12
QCD Approach Quarks Gluons
Quark Gluon Fragmentation Functions Q2
dependence predicted from QCD
FFF2 1978
Feynman quote from FFF2 We investigate whether
the present experimental behavior of mesons with
large transverse momentum in hadron-hadron
collisions is consistent with the theory of
quantum-chromodynamics (QCD) with asymptotic
freedom, at least as the theory is now partially
understood.
Parton Distribution Functions Q2 dependence
predicted from QCD
Quark Gluon Cross-Sections Calculated from QCD
13
High PT Jets
CDF (2006)
Feynman, Field, Fox (1978)
Predict large jet cross-section
30 GeV/c!
Feynman quote from FFF At the time of this
writing, there is still no sharp quantitative
test of QCD. An important test will come in
connection with the phenomena of high PT
discussed here.
600 GeV/c Jets!
14
A Parameterization of the Properties of Jets
Field-Feynman 1978
Secondary Mesons (after decay)

• Assumed that jets could be analyzed on a
  recursive principle.

• Let f(h)dh be the probability that the rank 1
  meson leaves fractional momentum h to the
  remaining cascade, leaving quark b with
  momentum P1 h1P0.

Rank 2
Rank 1

• Assume that the mesons originating from quark b
  are distributed in presisely the same way as the
  mesons which came from quark a (i.e. same
  function f(h)), leaving quark c with momentum
  P2 h2P1 h2h1P0.

Primary Mesons
continue

• Add in flavor dependence by letting bu
  probabliity of producing u-ubar pair, bd
  probability of producing d-dbar pair, etc.

Calculate F(z) from f(h) and bi!

• Let F(z)dz be the probability of finding a meson
  (independent of rank) with fractional mementum z
  of the original quark a within the jet.

Original quark with flavor a and momentum P0
15
Feynman-Field Jet Model
R. P. Feynman ISMD, Kaysersberg, France, June
12, 1977
Feynman quote from FF2 The predictions of the
model are reasonable enough physically that we
expect it may be close enough to reality to be
useful in designing future experiments and to
serve as a reasonable approximation to compare
to data. We do not think of the model as a
sound physical theory, ....
16
Monte-Carlo Simulationof Hadron-Hadron Collisions
FF1-FFF1 (1977) Black-Box Model
FF2 (1978) Monte-Carlo simulation of jets
F1-FFF2 (1978) QCD Approach
FFFW FieldJet (1980) QCD leading-log order
simulation of hadron-hadron collisions
FF or FW Fragmentation
the past
ISAJET (FF Fragmentation)
HERWIG (FW Fragmentation)
PYTHIA
today
tomorrow
SHERPA
PYTHIA 6.3
17
Monte-Carlo Simulationof Quark and Gluon Jets

• ISAJET Evolve the parton-shower from Q2 (virtual
  photon invariant mass) to Qmin 5 GeV. Use a
  complicated fragmentation model to evolve from
  Qmin to outgoing hadrons.

• HERWIG Evolve the parton-shower from Q2 (virtual
  photon invariant mass) to Qmin 1 GeV. Form
  color singlet clusters which decay into hadrons
  according to 2-particle phase space.

• MLLA Evolve the parton-shower from Q2 (virtual
  photon invariant mass) to Qmin 230 MeV. Assume
  that the charged particles behave the same as the
  partons with Nchg/Nparton 0.56!

Q2
MLLA Curve!
200 MeV
5 GeV
1 GeV
18
QCD Monte-Carlo ModelsHigh Transverse Momentum
Jets
Underlying Event

• Start with the perturbative 2-to-2 (or sometimes
  2-to-3) parton-parton scattering and add initial
  and final-state gluon radiation (in the leading
  log approximation or modified leading log
  approximation).

• The underlying event consists of the beam-beam
  remnants and from particles arising from soft or
  semi-soft multiple parton interactions (MPI).

The underlying event is an unavoidable
background to most collider observables and
having good understand of it leads to more
precise collider measurements!

• Of course the outgoing colored partons fragment
  into hadron jet and inevitably underlying
  event observables receive contributions from
  initial and final-state radiation.

19
Jet Algorithms

• Clustering algorithms are used to combine
  calorimeter towers or charged particles into
  jets in order to study the event topology and
  to compare with the QCD Monte-Carlo Models.

• We do not detect partons! The outgoing partons
  fragment into hadrons before they travel a
  distance of about the size of the proton. At
  long distances the partons manifest themselves as
  jets. The underlying event can also form
  jets. Most jets are a mixture of particles
  arising from the hard outgoing partons and the
  underlying event.

• Since we measure hadrons every observable is
  infrared and collinear safe. There are no
  divergences at the hadron level!

• Every jet algorithms correspond to a different
  observable and different algorithms give
  different results.

• Studying the difference between the algorithms
  teaches us about the event structure.

20
Jet Corrections Extrapolations

• Calorimeter Level Jets ? Hadron Level Jets
• We measure jets at the hadron level in the
  calorimeter.
• We certainly want to correct the jets for the
  detector resolution and efficiency.
• Also, we must correct the jets for pile-up.
• Must correct what we measure back to the true
  hadron level (i.e. particle level) observable!

Hadron ? Parton
I do not believe we should extrapolate the data
to the parton level! We should publish what we
measure (i.e. hadron level with the underlying
event)! To compare with theory we should
extrapolate the parton level to the hadron
level (i.e. add hadronization and the
underlying event to the parton level)! PYTHIA,
HERWIG, MC_at_NLO

• Particle Level Jets (with the underlying event
  removed)
• Do we want to make further model dependent
  corrections?
• Do we want to try and subtract the underlying
  event from the observed particle level jets.
• This cannot really be done, but if you trust the
  Monte-Carlo modeling of the underlying event
  you can do it by using the Monte-Carlo models
  (use PYTHIA Tune A).
• This is no longer an observable, it is a model
  dependent extrapolation!

Useless without a model of hadronization!

• Hadron Level Jets ? Parton Level Jets
• Do we want to use the data to try and extrapolate
  back to the parton level? What parton level,
  PYTHIA (Leading Log) or fixed order NLO?
• This also cannot really be done, but again if you
  trust the Monte-Carlo models you can try and do
  it by using the Monte-Carlo models (use PYTHIA
  Tune A) including ISR and FSR.
• Cannot extrapolate the data to fixed order NLO!

Next-to-leading order parton level calculation 0,
1, 2, or 3 partons!
21
Good and Bad Algorithms

• In order to correct what we see in the
  calorimeter back to the hadron level we must use
  an algorithm that can be defined at both the
  calorimeter and particle level.

• If you insist on extrapolating the data to the
  parton level then it is better to use an
  algorithm that is well defined at the parton
  level (i.e. infrared and collinear safe at the
  parton level).

• If you hadronize the parton level and add the
  underlying event (i.e. PYTHIA, HERWIG, MC_at_NLO)
  then you do not care if the algorithm is infrared
  and collinear safe at the parton level. You can
  predict any hadron level observable!

22
Four Jet Algorithms
Towers not included in a jet (i.e. dark towers)!
Bad

• JetClu is bad because the algorithm cannot be
  defined at the particle level.

• The MidPoint and Modified MidPoint (i.e. Search
  Cone) algorithms are not infrared and collinear
  safe at the parton level.

23
KT Algorithm

• kT Algorithm
• Cluster together calorimeter towers by their kT
  proximity.
• Infrared and collinear safe at all orders of
  pQCD.
• No splitting and merging.
• No ad hoc Rsep parameter necessary to compare
  with parton level.
• Every parton, particle, or tower is assigned to a
  jet.
• No biases from seed towers.
• Favored algorithm in ee- annihilations!

KT Algorithm
Will the KT algorithm be effective in the
collider environment where there is an
underlying event?
Raw Jet ET 533 GeV
Raw Jet ET 618 GeV
CDF Run 2
Only towers with ET gt 0.5 GeV are shown
24
KT Inclusive Jet Cross Section

• KT Algorithm (D 0.7)
• Data corrected to the hadron level
• L 385 pb-1
• 0.1 lt yjet lt 0.7
• Compared with NLO QCD (JetRad) corrected to the
  hadron level.

Sensitive to UE hadronization effects for PT lt
300 GeV/c!
25
Search Cone Inclusive Jet Cross Section

• Modified MidPoint Cone Algorithm (R 0.7, fmerge
  0.75)
• Data corrected to the hadron level and the parton
  level
• L 1.04 fb-1
• 0.1 lt yjet lt 0.7
• Compared with NLO QCD (JetRad, Rsep 1.3)

Sensitive to UE hadronization effects for PT lt
200 GeV/c!
26
Hadronization and Underlying Event Corrections
Note that DØ does not make any corrections for
hadronization or the underlying event!?

• Compare the hadronization and underlying event
  corrections for the KT algorithm (D 0.7) and
  the MidPoint algorithm (R 0.7)!

• We see that the KT algorithm (D 0.7) is
  slightly more sensitive to the underlying event
  than the cone algorithm (R 0.7), but with a
  good model of the underlying event both cross
  sections can be measured at the Tevatrun!

The KT algorithm is slightly more sensitive to
the underlying event!
27
KT Inclusive Jet Cross Section

• KT Algorithm (D 0.7).
• Data corrected to the hadron level.
• L 385 pb-1.
• Five rapidity regions
• yjet lt 0.1
• 0.1 lt yjet lt 0.7
• 0.7 lt yjet lt 1.1
• 1.1 lt yjet lt 1.6
• 1.6 lt yjet lt 2.1
• Compared with NLO QCD (JetRad) with CTEQ6.1

Excellent agreement over all rapidity ranges!
28
The Transverse Regionsas defined by the
Leading Jet
Charged Particle Df Correlations pT gt 0.5 GeV/c
h lt 1
Look at the charged particle density in the
transverse region!
Transverse region is very sensitive to the
underlying event!

• Look at charged particle correlations in the
  azimuthal angle Df relative to the leading
  calorimeter jet (JetClu R 0.7, h lt 2).
• Define Df lt 60o as Toward, 60o lt -Df lt 120o
  and 60o lt Df lt 120o as Transverse 1 and
  Transverse 2, and Df gt 120o as Away. Each
  of the two transverse regions have area DhDf
  2x60o 4p/6. The overall transverse region is
  the sum of the two transverse regions (DhDf
  2x120o 4p/3).

29
Run 1 PYTHIA Tune A
CDF Default!
PYTHIA 6.206 CTEQ5L
Parameter Tune B Tune A
MSTP(81) 1 1
MSTP(82) 4 4
PARP(82) 1.9 GeV 2.0 GeV
PARP(83) 0.5 0.5
PARP(84) 0.4 0.4
PARP(85) 1.0 0.9
PARP(86) 1.0 0.95
PARP(89) 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25
PARP(67) 1.0 4.0
Run 1 Analysis

• Plot shows the transverse charged particle
  density versus PT(chgjet1) compared to the QCD
  hard scattering predictions of two tuned versions
  of PYTHIA 6.206 (CTEQ5L, Set B (PARP(67)1) and
  Set A (PARP(67)4)).

Old PYTHIA default (more initial-state radiation)
Old PYTHIA default (more initial-state radiation)
New PYTHIA default (less initial-state radiation)
New PYTHIA default (less initial-state radiation)
30
Charged Particle Density Df Dependence
Refer to this as a Leading Jet event
Subset
Refer to this as a Back-to-Back event

• Look at the transverse region as defined by the
  leading jet (JetClu R 0.7, h lt 2) or by the
  leading two jets (JetClu R 0.7, h lt 2).
  Back-to-Back events are selected to have at
  least two jets with Jet1 and Jet2 nearly
  back-to-back (Df12 gt 150o) with almost equal
  transverse energies (ET(jet2)/ET(jet1) gt 0.8)
  and with ET(jet3) lt 15 GeV.

• Shows the Df dependence of the charged particle
  density, dNchg/dhdf, for charged particles in the
  range pT gt 0.5 GeV/c and h lt 1 relative to
  jet1 (rotated to 270o) for 30 lt ET(jet1) lt 70
  GeV for Leading Jet and Back-to-Back events.

31
Transverse PTsum Density vs ET(jet1)
Leading Jet
Back-to-Back
Min-Bias 0.24 GeV/c per unit h-f

• Shows the average charged PTsum density,
  dPTsum/dhdf, in the transverse region (pT gt 0.5
  GeV/c, h lt 1) versus ET(jet1) for Leading
  Jet and Back-to-Back events.

• Compares the (uncorrected) data with PYTHIA Tune
  A and HERWIG (without MPI) after CDFSIM.

32
TransMAX/MIN PTsum Density PYTHIA Tune A vs
HERWIG
PYTHIA Tune A does a fairly good job fitting the
PTsum density in the transverse region! HERWIG
does a poor job!
Back-to-Back
Leading Jet

• Shows the charged particle PTsum density,
  dPTsum/dhdf, in the transMAX and transMIN
  region (pT gt 0.5 GeV/c, h lt 1) versus PT(jet1)
  for Leading Jet and Back-to-Back events.
• Compares the (corrected) data with PYTHIA Tune A
  (with MPI) and HERWIG (without MPI) at the
  particle level.

33
TransMAX/MIN ETsum Density PYTHIA Tune A vs
HERWIG
Back-to-Back
Leading Jet
Neither PY Tune A or HERWIG fits the ETsum
density in the transferse region! HERWIG does
slightly better than Tune A!

• Shows the data on the tower ETsum density,
  dETsum/dhdf, in the transMAX and transMIN
  region (ET gt 100 MeV, h lt 1) versus PT(jet1)
  for Leading Jet and Back-to-Back events.
• Compares the (corrected) data with PYTHIA Tune A
  (with MPI) and HERWIG (without MPI) at the
  particle level (all particles, h lt 1).

34
TransDIF ETsum Density PYTHIA Tune A vs HERWIG
Leading Jet
Back-to-Back
transDIF is more sensitive to the hard
scattering component of the underlying event!

• Use the leading jet to define the MAX and MIN
  transverse regions on an event-by-event basis
  with MAX (MIN) having the largest (smallest)
  charged PTsum density.

• Shows the transDIF MAX-MIN ETsum density,
  dETsum/dhdf, for all particles (h lt 1) versus
  PT(jet1) for Leading Jet and Back-to-Back
  events.

35
Possible Scenario??

• PYTHIA Tune A fits the charged particle PTsum
  density for pT gt 0.5 GeV/c, but it does not
  produce enough ETsum for towers with ET gt 0.1 GeV.

• It is possible that there is a sharp rise in the
  number of particles in the underlying event at
  low pT (i.e. pT lt 0.5 GeV/c).

• Perhaps there are two components, a vary soft
  beam-beam remnant component (gaussian or
  exponential) and a hard multiple interaction
  component.

36
TransMAX/MIN ETsum Density PYTHIA Tune A vs
JIMMY
JIMMY was tuned to fit the energy density in the
transverse region for leading jet events!
JIMMY MPI J. M. Butterworth J. R. Forshaw M. H.
Seymour
Leading Jet
Back-to-Back

• Shows the ETsum density, dETsum/dhdf, in the
  transMAX and transMIN region (all particles
  h lt 1) versus PT(jet1) for Leading Jet and
  Back-to-Back events.
• Compares the (corrected) data with PYTHIA Tune A
  (with MPI) and a tuned version of JIMMY (with
  MPI, PTJIM 3.25 GeV/c) at the particle level.

37
TransMAX/MIN Nchg Density PYTHIA Tune A vs
JIMMY
Back-to-Back
Leading Jet

• Shows the charged particle density, dNchg/dhdf,
  in the transMAX and transMIN region (pT gt 0.5
  GeV/c, h lt 1) versus PT(jet1) for Leading
  Jet and Back-to-Back events.
• Compares the (corrected) data with PYTHIA Tune A
  (with MPI) and a tuned version of JIMMY (with
  MPI, PTJIM 3.25 GeV/c) at the particle level.

38
Transverse ltPTgt PYTHIA Tune A vs JIMMY
Back-to-Back
Leading Jet

• Shows the charged particle ltPTgt in the
  transverse (pT gt 0.5 GeV/c, h lt 1) versus
  PT(jet1) for Leading Jet and Back-to-Back
  events.
• Compares the (corrected) data with PYTHIA Tune A
  (with MPI) and HERWIG and a tuned version of
  JIMMY (with MPI, PTJIM 3.25 GeV/c) at the
  particle level.

Both JIMMY and HERWIG are too soft for pT gt 0.5
GeV/c!
39
QCD Monte-Carlo ModelsLepton-Pair Production
Underlying Event

• Start with the perturbative Drell-Yan muon pair
  production and add initial-state gluon radiation
  (in the leading log approximation or modified
  leading log approximation).

• The underlying event consists of the beam-beam
  remnants and from particles arising from soft or
  semi-soft multiple parton interactions (MPI).

• Of course the outgoing colored partons fragment
  into hadron jet and inevitably underlying
  event observables receive contributions from
  initial and final-state radiation.

40
The Central Regionin Drell-Yan Production
Look at the charged particle density and the
PTsum density in the central region!
Charged Particles (pT gt 0.5 GeV/c, h lt 1)
After removing the lepton-pair everything else is
the underlying event!

• Look at the central region after removing the
  lepton-pair.
• Study the charged particles (pT gt 0.5 GeV/c, h
  lt 1) and form the charged particle density,
  dNchg/dhdf, and the charged scalar pT sum
  density, dPTsum/dhdf, by dividing by the area in
  h-f space.

41
CDF Run 1 PT(Z)
PYTHIA 6.2 CTEQ5L
UE Parameters
Parameter Tune A Tune A25 Tune A50
MSTP(81) 1 1 1
MSTP(82) 4 4 4
PARP(82) 2.0 GeV 2.0 GeV 2.0 GeV
PARP(83) 0.5 0.5 0.5
PARP(84) 0.4 0.4 0.4
PARP(85) 0.9 0.9 0.9
PARP(86) 0.95 0.95 0.95
PARP(89) 1.8 TeV 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25 0.25
PARP(67) 4.0 4.0 4.0
MSTP(91) 1 1 1
PARP(91) 1.0 2.5 5.0
PARP(93) 5.0 15.0 25.0
ISR Parameter

• Shows the Run 1 Z-boson pT distribution (ltpT(Z)gt
  11.5 GeV/c) compared with PYTHIA Tune A
  (ltpT(Z)gt 9.7 GeV/c), Tune A25 (ltpT(Z)gt
  10.1 GeV/c), and Tune A50 (ltpT(Z)gt 11.2
  GeV/c).

Vary the intrensic KT!
Intrensic KT
42
CDF Run 1 PT(Z)
Tune used by the CDF-EWK group!
PYTHIA 6.2 CTEQ5L
Parameter Tune A Tune AW
MSTP(81) 1 1
MSTP(82) 4 4
PARP(82) 2.0 GeV 2.0 GeV
PARP(83) 0.5 0.5
PARP(84) 0.4 0.4
PARP(85) 0.9 0.9
PARP(86) 0.95 0.95
PARP(89) 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25
PARP(62) 1.0 1.25
PARP(64) 1.0 0.2
PARP(67) 4.0 4.0
MSTP(91) 1 1
PARP(91) 1.0 2.1
PARP(93) 5.0 15.0
UE Parameters
ISR Parameters

• Shows the Run 1 Z-boson pT distribution (ltpT(Z)gt
  11.5 GeV/c) compared with PYTHIA Tune A
  (ltpT(Z)gt 9.7 GeV/c), and PYTHIA Tune AW
  (ltpT(Z)gt 11.7 GeV/c).

Effective Q cut-off, below which space-like
showers are not evolved.
Intrensic KT
The Q2 kT2 in as for space-like showers is
scaled by PARP(64)!
43
Jet-Jet Correlations (DØ)

• MidPoint Cone Algorithm (R 0.7, fmerge 0.5)
• L 150 pb-1 (Phys. Rev. Lett. 94 221801 (2005))
• Data/NLO agreement good. Data/HERWIG agreement
  good.
• Data/PYTHIA agreement good provided PARP(67)
  1.0?4.0 (i.e. like Tune A, best fit 2.5).

44
CDF Run 1 PT(Z)
PYTHIA 6.2 CTEQ5L
Parameter Tune DW Tune AW
MSTP(81) 1 1
MSTP(82) 4 4
PARP(82) 1.9 GeV 2.0 GeV
PARP(83) 0.5 0.5
PARP(84) 0.4 0.4
PARP(85) 1.0 0.9
PARP(86) 1.0 0.95
PARP(89) 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25
PARP(62) 1.25 1.25
PARP(64) 0.2 0.2
PARP(67) 2.5 4.0
MSTP(91) 1 1
PARP(91) 2.1 2.1
PARP(93) 15.0 15.0
UE Parameters
ISR Parameters

• Shows the Run 1 Z-boson pT distribution (ltpT(Z)gt
  11.5 GeV/c) compared with PYTHIA Tune DW, and
  HERWIG.

Tune DW uses D0s perfered value of PARP(67)!
Intrensic KT
Tune DW has a lower value of PARP(67) and
slightly more MPI!
45
Transverse Nchg Density
PYTHIA 6.2 CTEQ5L
Three different amounts of MPI!
UE Parameters
Parameter Tune AW Tune DW Tune BW
MSTP(81) 1 1 1
MSTP(82) 4 4 4
PARP(82) 2.0 GeV 1.9 GeV 1.8 GeV
PARP(83) 0.5 0.5 0.5
PARP(84) 0.4 0.4 0.4
PARP(85) 0.9 1.0 1.0
PARP(86) 0.95 1.0 1.0
PARP(89) 1.8 TeV 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25 0.25
PARP(62) 1.25 1.25 1.25
PARP(64) 0.2 0.2 0.2
PARP(67) 4.0 2.5 1.0
MSTP(91) 1 1 1
PARP(91) 2.5 2.5 2/5
PARP(93) 15.0 15.0 15.0
ISR Parameter

• Shows the transverse charged particle density,
  dN/dhdf, versus PT(jet1) for leading jet
  events at 1.96 TeV for PYTHIA Tune A, Tune AW,
  Tune DW, Tune BW, and HERWIG (without MPI).

• Shows the transverse charged particle density,
  dN/dhdf, versus PT(jet1) for leading jet
  events at 1.96 TeV for Tune DW, ATLAS, and HERWIG
  (without MPI).

Three different amounts of ISR!
Intrensic KT
46
Transverse PTsum Density
PYTHIA 6.2 CTEQ5L
Three different amounts of MPI!
UE Parameters
Parameter Tune AW Tune DW Tune BW
MSTP(81) 1 1 1
MSTP(82) 4 4 4
PARP(82) 2.0 GeV 1.9 GeV 1.8 GeV
PARP(83) 0.5 0.5 0.5
PARP(84) 0.4 0.4 0.4
PARP(85) 0.9 1.0 1.0
PARP(86) 0.95 1.0 1.0
PARP(89) 1.8 TeV 1.8 TeV 1.8 TeV
PARP(90) 0.25 0.25 0.25
PARP(62) 1.25 1.25 1.25
PARP(64) 0.2 0.2 0.2
PARP(67) 4.0 2.5 1.0
MSTP(91) 1 1 1
PARP(91) 2.5 2.5 2/5
PARP(93) 15.0 15.0 15.0
ISR Parameter

• Shows the transverse charged PTsum density,
  dPT/dhdf, versus PT(jet1) for leading jet
  events at 1.96 TeV for PYTHIA Tune A, Tune AW,
  Tune DW, Tune BW, and HERWIG (without MPI).

• Shows the transverse charged PTsum density,
  dPT/dhdf, versus PT(jet1) for leading jet
  events at 1.96 TeV for Tune DW, ATLAS, and HERWIG
  (without MPI).

Three different amounts of ISR!
Intrensic KT
47
MIT Search Scheme 12
Exclusive 3 Jet Final State Challenge
CDF Data
At least 1 Jet (trigger jet) (PT gt 40 GeV/c,
h lt 1.0)
Normalized to 1
PYTHIA Tune A
Exactly 3 jets (PT gt 20 GeV/c, h lt 2.5)
R(j2,j3)
Order Jets by PT Jet1 highest PT, etc.
48
3Jexc R(j2,j3) Normalized
The data have more 3 jet events with small
R(j2,j3)!?

• Let Ntrig40 equal the number of events with at
  least one jet with PT gt 40 geV and h lt 1.0
  (this is the offline trigger).

• Let N3Jexc20 equal the number of events with
  exactly three jets with PT gt 20 GeV/c and h lt
  2.5 which also have at least one jet with PT gt 40
  GeV/c and h lt 1.0.

Normalized to N3JexcFr

• Let N3JexcFr N3Jexc20/Ntrig40. The is the
  fraction of the offline trigger events that are
  exclusive 3-jet events.

• The CDF data on dN/dR(j2,j3) at 1.96 TeV compared
  with PYTHIA Tune AW (PARP(67)4), Tune DW
  (PARP(67)2.5), Tune BW (PARP(67)1).

• PARP(67) affects the initial-state radiation
  which contributes primarily to the region
  R(j2,j3) gt 1.0.

49
3Jexc R(j2,j3) Normalized
I do not understand the excess number of
events with R(j2,j3) lt 1.0. Perhaps this is
related to the soft energy problem?? For now
the best tune is PYTHIA Tune DW.

• Let Ntrig40 equal the number of events with at
  least one jet with PT gt 40 geV and h lt 1.0
  (this is the offline trigger).

• Let N3Jexc20 equal the number of events with
  exactly three jets with PT gt 20 GeV/c and h lt
  2.5 which also have at least one jet with PT gt 40
  GeV/c and h lt 1.0.

Normalized to N3JexcFr

• Let N3JexcFr N3Jexc20/Ntrig40. The is the
  fraction of the offline trigger events that are
  exclusive 3-jet events.

• The CDF data on dN/dR(j2,j3) at 1.96 TeV compared
  with PYTHIA Tune DW (PARP(67)2.5) and HERWIG
  (without MPI).

• Final-State radiation contributes to the region
  R(j2,j3) lt 1.0.

• If you ignore the normalization and normalize all
  the distributions to one then the data prefer
  Tune BW, but I believe this is misleading.

50
Drell-Yan Production (Run 2 vs LHC)
Lepton-Pair Transverse Momentum
ltpT(mm-)gt is much larger at the LHC!
Shapes of the pT(mm-) distribution at the
Z-boson mass.
Z

• Average Lepton-Pair transverse momentum at the
  Tevatron and the LHC for PYTHIA Tune DW and
  HERWIG (without MPI).

• Shape of the Lepton-Pair pT distribution at the
  Z-boson mass at the Tevatron and the LHC for
  PYTHIA Tune DW and HERWIG (without MPI).

51
The Underlying Event inDrell-Yan Production
The Underlying Event
Charged particle density versus M(pair)
HERWIG (without MPI) is much less active than PY
Tune AW (with MPI)!
Underlying event much more active at the LHC!
Z
Z

• Charged particle density versus the lepton-pair
  invariant mass at 1.96 TeV for PYTHIA Tune AW and
  HERWIG (without MPI).

• Charged particle density versus the lepton-pair
  invariant mass at 14 TeV for PYTHIA Tune AW and
  HERWIG (without MPI).

52
Extrapolations to the LHCDrell-Yan Production
Charged particle density versus M(pair)
The Underlying Event
Tune DW and DWT are identical at 1.96 TeV, but
have different MPI energy dependence!
Z
Z

• Average charged particle density versus the
  lepton-pair invariant mass at 1.96 TeV for PYTHIA
  Tune A, Tune AW, Tune BW, Tune DW and HERWIG
  (without MPI).

• Average charged particle density versus the
  lepton-pair invariant mass at 14 TeV for PYTHIA
  Tune DW, Tune DWT, ATLAS and HERWIG (without
  MPI).

53
Extrapolations to the LHCDrell-Yan Production
Charged particle charged PTsum density versus
M(pair)
The Underlying Event
The ATLAS tune has a much softer distribution
of charged particles than the CDF Run 2 Tunes!
Z
Z

• Average charged PTsum density versus the
  lepton-pair invariant mass at 14 TeV for PYTHIA
  Tune DW, Tune DWT, ATLAS and HERWIG (without
  MPI).

• Average charged PTsum density versus the
  lepton-pair invariant mass at 1.96 TeV for PYTHIA
  Tune A, Tune AW, Tune BW, Tune DW and HERWIG
  (without MPI).

54
Extrapolations to the LHCDrell-Yan Production
Charged particle density versus M(pair)
The Underlying Event
The ATLAS tune has a much softer distribution
of charged particles than the CDF Run 2 Tunes!
Charged Particles (hlt1.0, pT gt 0.5 GeV/c)
Charged Particles (hlt1.0, pT gt 0.9 GeV/c)
Z
Z

• Average charged particle density (pT gt 0.5 GeV/c)
  versus the lepton-pair invariant mass at 14 TeV
  for PYTHIA Tune DW, Tune DWT, ATLAS and HERWIG
  (without MPI).

• Average charged particle density (pT gt 0.9 GeV/c)
  versus the lepton-pair invariant mass at 14 TeV
  for PYTHIA Tune DW, Tune DWT, ATLAS and HERWIG
  (without MPI).

55
Constraining the Higgs Mass

• Top quark mass is a fundamental parameter of SM.
• Radiative corrections to SM predictions dominated
  by top mass.
• Top mass together with W mass places a constraint
  on Higgs mass!

Summer 05
Light Higgs very interesting for the Tevatron!
56
20 Years of Measuring W Z
57
The WW Cross Section
Campbell Ellis 1999
pb-1 CDF (pb) NLO (pb)
s(WW) CDF 184 14.65.8(stat)-5.1(stat)?1.8(sys)?0.9(lum) 12.4?0.8
s(WW) DØ 240 13.84.3(stat)-3.8(stat)?1.2(sys)?0.9(lum) 12.4?0.8
58
The WW Cross Section
L 825 pb-1

• WW?dileptons MET
• Two leptons pT gt 20 GeV/c.
• Z veto.
• MET gt 20 GeV.
• Zero jets with ETgt15 GeV and hlt2.5.

We are beginning to study the details
of Di-Boson production at the Tevatron!
Observe 95 events with 37.2 background!
L CDF (pb) NLO (pb)
s(WW) 825 pb-1 13.7?2.3(stat)?1.6(sys)?1.2(lum) 12.4?0.8
Missing ET!
Lepton-Pair Mass!
ET Sum!
59
Di-Bosons at the Tevatron
W
We are getting closer to the Higgs!
Z
Wg
Zg
WW
WZ
60
The Z?tt Cross Section

• Taus are difficult to reconstruct at hadron
  colliders
• Exploit event topology to suppress backgrounds
  (QCD Wjet).
• Measurement of cross section important for Higgs
  and SUSY analyses.
• CDF strategy of hadronic t reconstruction
• Study charged tracks define signal and isolation
  cone (isolation require no tracks in isolation
  cone).
• Use hadronic calorimeter clusters (to suppress
  electron background).
• p0 detected by the CES detector and required to
  be in the signal cone.
• CES resolution 2-3mm, proportional strip/wire
  drift chamber at 6X0 of EM calorimeter.

• Channel for Z?tt electron isolated track
• One t decays to an electron t?eX (ET(e) gt 10
  GeV) .
• One t decays to hadrons t ? hX (pT gt 15GeV/c).
• Remove Drell-Yan ee- and apply event topology
  cuts for non-Z background.

61
The Z?tt Cross Section

• CDF Z?tt (350 pb-1) 316 Z?tt candidates.
• Novel method for background estimation main
  contribution QCD.
• t identification efficiency 60 with
  uncertainty about 3!

CDF (pb) NNLO (pb)
s(Z?tt-) 265?20(stat)?21(sys)?15(lum) 252.3?5.0
62
Higgs ? tt Search
events
Lets find the Higgs! Higgs Discovery Group
140 GeV Higgs Signal!
1 event

• Data mass distribution agrees with SM
  expectation
• MH gt 120 GeV 8.40.9 expected, 11 observed.
• Fit mass distribution for Higgs Signal (MSSM
  scenario)
• Exclude 140 GeV Higgs at 95 C.L.
• Upper limit on cross section times branching
  ratio.

63
Job Searching Craig Group

• Craig Group will graduate in December 2006 and is
  looking for a postdoctoral position.
• He is a student of myself and K. Matchev
  (phenomenology).
• His CDF thesis is the 1 fb-1 jet cross section
  (central and forward).
• He was one of the authors on LHAPDF.
• He set a CDF-CAF at Florida.
• He is good at both theory and analysis.
• He would like to continue to work on CDF for
  several years and then move to the LHC.
• He is an excellent physicist! One of the best
  students I have had!