Comparison HARP data with GEANT4 and MARS Simulations

1
Comparison HARP data with GEANT4 and MARS
Simulations
NUFACT06, UC Irvine 28 August 2006 Stephen
Brooks1, Paul Soler2, Kenny Walaron2
1Rutherford Appleton Laboratory 2University of
Glasgow/RAL
2
Motivation

• Determination of proton energy at a Neutrino
  Factory needs simulations of protons on different
  targets
• Mars 14 predicts broad spectrum above 5 GeV/c

? 6 11 GeV
? 9 19 GeV
We adopted 10 5 GeV as representative range
Zisman, ISS Accelerator WG summary
3
Motivation

• Need to validate hadronic models for neutrino
  target simulations
• There were comparisons of Geant4 and Mars15
  target simulations reported at NUFACT05 (Brooks,
  Walaron)
• Problem no good agreement amongst different
  hadronic models of Geant4 and Mars.

Pion yields passing through front-end of Neutrino
Factory
Geant4 QGSP model
4
Motivation

• Geant 4 secondaries after protons impinging on
  target

• Fate of Geant 4 secondaries at a Neutrino
  Factory front-end

5
Geant4/MARS Hadronic Models

• Geant4 hadronic models (see G4 Physics Reference
  Manual) R. Raja talk plenary
• LHEP GHEISHA parametrization used in GEANT3,
  normally valid Elt25 GeV
• LHEP-BERT Elt3 GeV Bertini intranuclear cascade,
  Egt3 GeV GHEISHA
• LHEP-BIC Elt3 GeV Binary intranuclear cascade,
  Egt3 GeV GHEISHA
• QGSP Elt25 GeV GHEISHA, Egt25 GeV quark-gluon
  string pre-compound
• QGSP-BERT Elt3 GeV Bertini cascade, 3ltElt25 GeV
  GHEISHA, Egt25 GeV
• quark-gluon string pre-compound model
• QGSP-BIC Elt3 GeV Binary cascade, 3ltElt25 GeV
  GHEISHA, Egt25 GeV
• quark-gluon string model
• QGSC Elt25 GeV GHEISHA, Egt25 GeV quark-gluon
  string with CHIPS
• (Chiral invariant Phase Space) model
• MARS 15 hadronic model
• Cascade-Exciton Model for Elt5 GeV
• Inclusive hadron production parametrisation for
  Egt5 GeV
• Need to establish validity of models by
  contrasting to HARP data

6
HARP Experiment

• HARP investigation of pion and kaon
  cross-sections from protons impinging on a
  variety of targets between 1.5-15 GeV/c

Described D Schmitz, S Borgi talks
7
HARP Data

• HARP results so far
• Aluminium target (5LI) at 12.9 GeV/c (K2K
  target) in forward direction
• published Nucl.Phys.B7321-45,2006
• Beryllium target at 8.9 GeV/c (MiniBooNE) and
  carbon at 12 GeV/c in forward direction see
  David Schmitzs talk in plenary session
• Large angle tantalum target at 3, 5, 8 GeV/c
• see Silvia Borghis talk in WG3
• I will not describe analysis. Refer to papers and
  other talks.
• This talk is comparison of HARP 12.9 GeV/c
  cross-section data to cross-section predictions
  from Geant4 and MARS models.

HARP 5 LI 12.9 GeV data
8
HARP Data

• Differential cross-section definition used in
  HARP analysis
• Number protons
• Target nuclei per unit area
• Number of observed particles of type a in
  momentum bin pi and angle qj
• Inverse of correction matrix that corrects for
  detector efficiency and resolution
• For simulated data, we only need
• (MC truth makes effic. matrix unitary)

• Monte Carlo samples 5x106 pot at 5 LI and 5x106
  pot at 1 LI
• Linear extrapolation to 0 LI corrects for
  re-absorption in target.

9
HARP Data

• Differential cross-sections

10
Comparison G4 with HARP

• Comparison Geant4 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Deficit HARP 1.0-2.0 GeV/c
30-60 mrad
60-90 mrad
11
Comparison G4 with HARP

• Comparison Geant4 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

90-120 mrad
120-150 mrad
Deficit HARP 1.0-1.5 GeV/c
12
Comparison G4 with HARP

• Comparison Geant4 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

150-180 mrad
180-210 mrad
Deficit HARP 1.0 GeV/c
13
Comparison G4 with HARP

• Comparison Geant4 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Deficit HARP 1.0-2.0 GeV/c
Total cross-section vs momentum
Total cross-section vs angle
14
Comparison G4 with HARP

• However, can tune parameters in Geant 4 (e.g.
  tune of QGSC model)

V. Ivantchenko
15
Comparison MARS 15 with HARP

• Comparison MARS 15 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Deficit MARS 1.0-2.0 GeV/c
16
Comparison MARS 15 with HARP

• Comparison MARS 15 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Good agreement
17
Comparison MARS 15 with HARP

• Comparison MARS 15 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Good agreement
18
Comparison MARS 15 with HARP

• Comparison MARS 15 with HARP 12.9 GeV/c Aluminium
  data (5 LI) as a function of angle
• Forward direction 30-210 mrad

Deficit HARP 1.0-2.0 GeV/c
Total cross-section vs momentum
Total cross-section vs angle
19
Comparison MARS/Geant4 with HARP

• MARS differences with HARP data smaller (lt40)
  than with G4-QGSC (100 at low momentum, low
  angle)

MC-HARP HARP
20
Outlook

• We have compared HARP data to GEANT4 and MARS15
• Deficit of HARP data with respect to Geant 4 and
  MARS simulations between 1.0-2.0 GeV/c for low
  angle
• MARS simulation has better agreement than best
  GEANT4 simulation (QGSC)
• However, can tune parameters of G4-QGSC to obtain
  better agreement.
• Still need to include FLUKA as another model for
  comparison
• As HARP data becomes available, there will be a
  comprehensive programme of comparison of MC
  models with all data and to phenomenological
  models