Inclusive jet photoproduction at HERA

1
Inclusive jet photoproduction at HERA
2
Jet photoproduction (parton level)
3
Jet photoproduction (hadron level)
LO QCD partons ? jets of hadrons ? detector
signals
LO hard process
Soft processes
Higher order QCD processes

• Compare jets at the parton, hadron and detector
  level
• Jet algorithms must ensure
• infrared and collinear safety
• minimal sensitivity to non-perturbative processes
• inclusive k? (D1) algorithm (EllisSoper,
  PRD48 (1993) 3160)
• dij min (ETi ,ETj ) Rij /D ET weighted
  recombination scheme
• Cone (R1) algorithm used for comparison with
  previous data

4
Motivation

• High ET jets (? non-perturbative effects and
  scale uncertainty reduced)
• Direct insight into parton dynamics
• Precise tests of perturbative QCD predictions
• Constrain photon and proton PDFs
• Search for new physics
• Low ET jets (? non-perturbative effects and
  scale uncertainty important)
• Test phenomenological models of underlying
  event fragmentation
• Inclusive vs dijet
• More statistics, extended kinematical range
• ? No direct reconstruction of xg ,xp
• No infrared sensitivity w.r.t. kinematical cuts
  as for dijet

5
QCD calculations and Monte Carlo

• Most precise QCD calculations up to NLO (parton
  level)
• NLO QCD weighted parton Monte Carlo (Frixione,
  NPB507(1997) 295)
• Photon proton PDFs GRV CTEQ5M
• Other choice photon ? AFG
• proton ? MRST99, CTEQ5HJ
  (enhanced gluon at high xp )
• LO QCD Monte Carlo event generators to correct
  data and calculations to the hadron level
  PYTHIA, PHOJET, HERWIG
• LQCD 200 MeV
• Fragmentation LUND String (PYTHIA, PHOJET) or
  Cluster (HERWIG)

6
H1 detector at HERA
7
Experimental facts (I)

• Inclusive cross section as a function of ETjet
  and hjet
• count the number of jets in a given kinematical
  range
• hjet measured in laboratory frame (hcms hjet
  2)
• High ET jets (ETjet gt 21 GeV)
• L 24 pb-1, untagged (e undetected) data
• ? Q2 lt 1 GeV2 , 95 lt W gp lt 285 GeV (0.1 lt y
  lt 0.9)
• Low ET jets (5 GeVlt ETjet lt 21 GeV)
• L 0.5 pb-1, tagged (e detected) data
• ? Q2 lt 0.01 GeV2 , 164 lt W gp lt 242 GeV (0.3
  lt y lt 0.65)

8
Experimental facts (II)
Hadron level cross sections obtained using Monte
Carlo

• Hadronisation corrections
• fragmentation (after parton showers)
• underlying event (after fragmentation)
• reverse order ? consistent results
• (1dhadr.) (1dfrag.) . (1du.e.)
• dfrag.lt 0 and? when ET ? or h?
• du.e.gt 0 and? when ET ? or h?
• dhadr. 30 (10) for ET lt10 (gt20) GeV
• Cone dhadr. 40 (20) for ET lt15 (gt15) GeV

• Data corrections
• bin migrations
• Important due to
• steeply falling ET spectrum
• selection efficiencies
• Exclude regions of large migrations
• high ET
• hlt0 (photon region)
• low ET
• hgt1.5 (proton region)

9
Experimental facts (III)

• Systematic uncertainties
• LAr hadronic energy scale ? 10-20 (10 ) for
  low (high) ET
• Correction for detector effects ? lt 10 (8 )
  for low (high) ET (statistical ?
  1/2 difference between Monte Carlo)
• Luminosity ? 1.5
• All other uncertainties (SPACAL energy scale,
  fraction of hadronic energy flow carried by
  tracks, background subtraction, trigger
  efficiency) ? 1
• Theoretical uncertainties
• Hadronisation correction uncertainty ? 30 (10
  ) for low (high) ET (statistical ? 1/2
  difference between Monte Carlo)
• Renormalisation factorisation scale (x2, /2)
  uncertainty ? lt 10

10
ET h distribution (high ET)

• LO too low at low ET and high h
• Agreement with NLO very good, even w/o
  hadronisation corrections
• All predictions using different PDFs agree with
  the data

11
ET distribution (W gp bins, high ET)
ET h fixed ltxg ,xpgt ? 1/W gp

• LO prediction
• low ET high W gp
• too low
• NLO prediction
• high W gp
• very good agreement
• lowW gp
• reasonable agreement
• promising region to
• constrain gluon at high xp

12
ET distribution full range

• LO prediction fails to reproduce shape
• NLO prediction
• good agreement over 6 orders of
  magnitude!
• hadronisation corrections needed
• Fit
• Range 5 lt ET lt 35 GeV
• n 7.5 ? 0.3 (stat) 0.1 0.5 (syst.)
• compatible with similar fit on charged
  particle cross section
• (EPJ C10 (1999) 363)
• n 7.03 ? 0.07 -0.2 (syst.)

13
h distribution (ET bins, high ET)
W gp h fixed ltxg ,xpgt ? ET

• Good agreement with NLO even w/o hadronisation
  corrections
• Precision of data equivalent to (or even better
  than) scale uncertainty
• challenge for theory to reduce uncertainty

14
h distribution (ET W gp bins, high ET)
h fixed ltxg ,xpgt ? ET / W gp

• Good agreement with NLO even w/o hadronisation
  corrections
• Cross section maximum shifted towards lower h
  values for higher W gp (Lorentz boost) and lower
  ET
• All PDFs consistent with data
• Precision of data equivalent to (or better
  than) scale uncertainty
• could be used to better constrain PDFs fits

15
h distribution (ET bins, low ET)

• 12 lt ET lt 21 GeV
• good agreement
• both NLO and hadronisation corrections needed
• 5 lt ET lt 12 GeV
• data indicative of a trend different from
  calculation
• challenge for Monte Carlo to accurately
  estimate hadronisation corrections?
• inadequacy of photon PDFs?
• higher-order terms needed?

16
Comparison with pp
-
Scaled cross section (independent of energy up to
scaling violations)

• xT lt 0.2
• shape similar for gp and pp
• resolved photon hadron
• xT gt 0.2
• gp harder than pp spectrum
• enhanced quark density in the resolved photon
  w.r.t. a hadron
• dominance of direct
• point-like photon

-
-

Confirmation of the dual nature of the photon
17
Summary

• New measurement of inclusive jet photoproduction
  cross section (L x 80 compared with previous one)
  using the k? algorithm
• Kinematical range extended to ET 75 GeV ( xT
  0.5)
• Experimental uncertainties already competitive
  with (scale) uncertainties
• Good agreement over 6 orders of magnitude in ET
  distribution
• NLO and hadronisation corrections needed,
  especially at low ET
• No discrimination of PDFs, but data helpful in
  global PDFs fits and future measurement promising
  for the gluon at high xp
• Determination of hadronisation corrections
  challenging for theory and phenomenology
• Comparison of scaled cross section with pp data
  confirms the dual nature of the photon with a
  transition around xT 0.2

-