Detector Simulation (Geant4) Activities at Fermilab

1
Detector Simulation (Geant4) Activities at
Fermilab V. Daniel Elvira CD Project Report
2
Outline

• Recent history on Fermilab contributions to
  (Geant4) detector simulation
• PSM Current Activities on Detector Simulations
• LHC Physics Center (LPC).
• CMS Experiment.
• Fermilab Users Support (J. Yarbas talk
  next).
• Fermilab Proposal to the Geant4 Collaboration
• Tool kit development projects by PSM core
  group.
• Collaboration with other Fermilab
  interested parties.

3
PSM and Detector Simulations
Mission support, stimulate, initiate simulation
efforts associated with Fermilab programs
develop detector simulation expertise, including
the use of software tools (nowadays mainly
Geant4), among Fermilab users.
What is Geant4 ?
A toolkit for the simulation of the passage of
particles through matter and EM fields. It is the
C OO descendant of Geant3. ( Collaboration
of more than 100 physicists computing
professionals CERN, SLAC,, universities )
Fermilab is not yet a collaborating
institution, a situation that partially limits
the fulfillment of our mission.
4
Recent History at the Lab
Until early 2005, Fermilab had made isolated
but significant contributions to validation,
support, development. Mostly through scientists
working on a given project or experiment.

• Implementation of time dependent EM fields in
  Geant4, visualization
• Beam Tools to simulate accelerator elements
  cooling simulations for Neutrino Source and Muon
  Collider feasibility studies.
• CMS Hcal Test Beam simulation and hadronic
  physics validation
• MiniBoone beam line, BTeV simulation framework,
  applications, and tests
• Local installation of Geant4 releases twice a
  year
• Three days tutorials in 2003, general support to
  FNAL users

People involved D. Elvira, L. Garren, J.
Kallenbach, P. Lebrun,
P. Spentzouris, J. Yarba
5
Current PSM Group Activities
Geant4 is now widely used in current and future
HEP projects associated with Fermilab CMS,
Neutrino Astrophysics Experiments,
International Linear Collider detector studies,
Neutron Therapy Facility
Decision made to build a Geant4 core group
within PSM
Daniel Elvira (scientist) 85 simulation
activities (down to 50 in 07) Julia Yarba (CP)
100 simul. activities Sunanda Banerjee coming
Sept06 (App. Phys.) 100 simul. activities
In 05-06 improve general simulation support
(see Julias talk), support/participation USCMS
simulation activities at the LPC, ILC-SLIC
releases (L. Garren). From 06 on expand
general simulation support, continue LPC/CMS
activities, participate of G4 development as
collaborating institution
6
The LPC Simulations Group
LHC Physics Center created to provide USCMS
collaborators with a physical place to work in
their own time zone where they will find a
remote control room, excellent video meeting
resources, computing infrastructure, software
expertise, and all elements to do physics in a
collaborative environment. LPC the US CMSD0
equivalent of the D0 or CDF trailers.
A virtual building 40 this side of the
Atlantic Simulations Group H. Cheung, D.
Elvira (Leaders), Julia Yarba (support,
infrastructure software development, code
integration) Provide support, generate expertise
to help USCMS institutions to carry out their own
simulation projects
Participating Institutions in LPC-Simulations
Brown, Carnegie-Melon, Colorado, FNAL, FSU, KSU,
KU, Maryland, Mississippi, Northwestern, Notre
Dame, Purdue-Calumet, Rutgers, Rockefeller,
SUNY-Buffalo, UIC (USA)
20 people, 70 in mailing
list
7
LPC-Simulation Activities

• Luminosity Monitoring System Simulation
• ECAL Crystal Response Studies
• Zero Degree Calorimeter (ZDC) in Simulation
• Forward Pixels Geometry and Simulation
• Geant4 physics validation using TB data
• Simulation Validation Suite
• GFlash parameterization of hadronic showers

Large and growing fraction of CMS simulation
activities
Julia Yarba is a key contributor to
infrastructure development (particle gun, vertex
smearing), integration (simulation chain,
validation suite), testing (debugging timing
studies), and support (questions,
tutorials) Responsibilities in international
CMS simulation group co-coordination Physics
Software Validation group coordination (D.
Elvira).
If you are a member of international CMS and work
on simulations, the LPC is the best place to be
8
The Phys. Soft. Validation Suite

• Detect automatically and correct problems in new
  release with respect to a previous reference.
• The LPC developed the Simulation Validation
  Suite (G4 hits) late 2005
• Charge to the PSV Group Develop Physics
  Software Validation Suite, for hits, digis,
  rechits, physics objects for CMSSW.

Validation branch (Pool files, ROOT browsable)
Pre-generated samples single particles, pp
physics (Pool files, ROOT browsable, HepMC
format)
PSVS software sub-systems (dedicated
packages)
Packages currently available Performance,
Mixing,TrackerHits, TrackerDigis, EcalHits,
EcalDigis, HcalHits, HcalDigis, Geometry,
GlobalHits, TrackerRecHits, DTMuonsRecHits
OVAL
Configuration file commands, tolerance values
9
PSVS People Documentation
USCMS Institutions and People Involved with PSVS
S. Abdullin, H. Cheung, D. Elvira, J. Yarba
FNAL X. Huang -
Mayaguez (Puerto Rico), A. Kharchilava, M.
Strang -SUNY Buffalo, L. Shabalina - UIC D.
Andulin Virginia many more from international
CMS
Status first integrated version of the PSVS to
be commissioned with CMSSW_07_0 by mid-June 2006
10
HCAL TB04 Data/MC Comparison
Final data analysis results from HCAL TB 2004
from J. Damgov, compared with Geant4
predictions by S. Piperov, based on
OSCAR/G4.6.2.p02. Part of PTDR and shown in
LCG-Simulation Validation
VLE beam (ATLAS result not available yet)
Plan to compare this data with G4.7.1 and G4.8.0
(improved hadronic shower)
11
HCAL TB04 Data/MC Comparison
TB06 this summer better particle ID for VLE
beams.
12
Forward Pixels Simulation
Activity initiated at the LPC in March 2005. Now
these team is a main player in the pixels
simulation
People Involved Victoria Martin (NW) wrote
plaquette geometry Dmitri Onoprienko (KS) main
geometry developer (DDD expert) Xingtao Huang
(Mayaguez/Pto.Rico) implemented numbering
scheme, simulation under CMSSW, digi
validation. Colorado group - geometry of pixel
service cylinder, partial pixel test 07 Neeti
Parashar (LTU-Calumet) geometry, hits,
validation representative to simulation and
validation groups.
Status simulation operational under CMSSW,
geometry re-written to reflect changes in
engineering drawings. Validation packages ready
for integration to the suite.
13
Forward Pixels Simulation (II)
Disks
Plaquette
Pannel
14
Zero Degree Calorimeter (ZDC)
Cerenkov detector located 140 m from the
interaction point to measure forward g, n within
the Heavy Ion Program
People Involved Michael Murray-Project Manager
(KU) Megan Lehnherr
(KU), Chadd Smith (UIC) in OSCAR
Edmundo Garcia (UIC) ported to CMSSW
Status Edmundo Garcia recently finished the
porting the standalone ZDCTest
simulation to CMSSW, fixed bugs, validate physics
of G4 using RIHC test
beam experiment. Next digitization, shower
parameteriz.
of photons per fiber in layers of Hadronic
section of ZDC 1000 GeV neutron 500 GeV neutron
15
GFlash Parameteriz. of Hadronic Showers

• GFlash is a package (developed by H1) to
  parameterize particle
• showers in a MC simulation. We could gain
• Speed, but this is not main goal.
• Accuracy, by tuning it to data in case that the
  MC is not accurate enough.
• The Geant4 team is working hard to understand
  underlying physics processes in the particle
  shower, and provide a most accurate product
  possible given the current physics knowledge.
• GFlash can help us take the additional step
  towards final tuning of our detector simulation
  to test beam and collider data.
• Long Term Goal accurate (and fast enough) MC to
  derive, understand, estimate corrections,
  systematic uncertainties in data analysis.

USCMS (LPC) Team R. Godang R. Kroeger
(Mississippi), Soon Jun M. Paulini
(Carnegie-Mellon), Anwar Bhatti (Rockefeller)
others from international CMS
16
The Future

• With very little man-power our proto-Geant4 core
  group initiated,
• enabled, and currently supports a significant
  amount of simulation activity
• Related to Fermilab research programs.
• Where/how should be expand?
• Should we become members of the Geant4
  Collaboration?
• Number 1 is an open question ILC, neutrino,
  astrophysics experiments,
• YES to 2. The high level of expertise needed to
  pursue our goals can only
• be achieved through direct participation in the
  simulation toolkit
• development team. Collaboration membership would
  also enable Fermilab to
• have a say in the future evolution of the
  project, as well as improve
• communication in both directions.

Fermilab Proposal to the Geant4 Collaboration in
preparation
17
The Proposal
Who? The members of the PSM Geant4 core group
(Julia, Sunanda, Daniel) all who are willing to
do work would be members of the G4 Collab. Other
interested parties will be invited to sign the
proposal in support. Conditions presentation of
proposal (no FTE requirement). But 2 FTE
needed for representation in the Steering Board
(direction of the Collaboration) What? Hadronic
Physics. An area of significant interest for
Fermilab programs and scientists hadron
calorimetry at the ILC and LHC, neutrino fluxes
and atmospheric showers. An area where the G4
team is in need for help. For How Long? For as
long as Geant4 is relevant to the Fermilab
programs or all related issues of interest have
been addressed/solved.
18
Hadronic Physics
Physics modeling of hadronic interations for
ranges from thermal energies for neutron
cross-sections, through 7 TeV (LHC), to higher
energies in cosmic ray physics. Interaction
Cross-Sections and Modeling of final
states measured data, parameterizations/extrapola
tions based on experimental data under
theoretical assumptions, or theory based. Example
s (physics sets or lists) LHEP model based
on data carried over from Geant4 (GHEISHA) QGSP
based on Quark Gluon String theoretical model for
high Energy hadron-nucleus inelastic scattering.
19
Hadronic Physics Tasks

• Modeling of transition regions, connecting high
  E (theory based) with low E (data based) 8-20
  GeV. Validity of QGS models at high energy.
• Validation. Connect macroscopic quantities (such
  as shower shapes) with Geant4 physics processes
  and identify the problems.

Current issue is that showers too narrow and
short at high E. Investigating, by particle type,
contributions from process cross-section, neutron
production, energy spectrum at different depths,
etc. Current LHC TB programs useful tool for
validation.

• Others Alternate models.
• Had. shower parameterization
  infrastructure,

Iterating with G4 Collaboration
20
Summary
With very little man-power our proto-Geant4 core
group initiated, enabled, and currently supports
a significant amount of simulation activity
related to Fermilab research programs. Follow
model of creating user based teams for relevant
projects. Ready to take next step Geant4
toolkit development Opportunities to collaborate
with Fermilab user institutions, eventually
external funding.