Simulation Software for GLAST

1
Simulation Software for GLAST

• Alessandro de Angelis
• University of Udine, INFN Trieste and LIP Lisboa
• Round Table on 21st Century MC Methods for Space
  Applications
• 13-15 June 2001
• ESTEC, Noordwijk, The Netherlands

2
The cast

• Italy (5 FTE)
• Udine (2 students, 2 computer scientists, 1
  postdoc, 2 staff)
• Trieste/Ferrara (Francesco)
• Perugia (1 student, 1 staff)
• Bari (1 PhD)
• Roma2 (dark matter simulation, 1 staff 2
  students) and Pisa (digitizations, 1 engineer and
  2 staff)
• Japan (quicktest for the balloon flight)
• 2 students FTE from Hiroshima 1 staff from Tokio

3
GLAST
Tracker

• g telescope on satellite for the range 20 MeV-300
  GeV
• hybrid tracker calorimeter
• International collaboration NASA
  US-France-Italy-Japan-Sweden
• Broad experience in high-energy astrophysics and
  particle physics (science instrumentation)
• Timescale 2006-2010 (-gt2015)
• Wide range of physics objectives
• Gamma astrophysics
• Fundamental physics

Calorimeter
A HEP / astrophysics partnership
4
GLAST the instrument

• Tracker
• Si strips converter
• Calorimeter
• CsI with diode readout
• (a classic for HEP)
• 1.7 x 1.7 m2 x 0.8 m
• height/width 0.4 ? large FOV
• 16 towers ? modularity

5
GLAST the tracker

• Si strips converter
• High signal/noise
• Rad-hard
• Low power
• 4x4 towers, of 37 cm ? 37 cm of Si
• 18 x,y planes per tower
• 19 tray structures
• Electronics on the sides of trays
• Minimize gap between towers
• Carbon-fiber walls to provide stiffness

6
GLAST The calorimeter
CsI with diode readout

• Good E resolution
• High signal/noise
• Hodoscope good position determination leakage
  correction
• 4x4 arrays of CsI (Tl) crystals
• Thickness of 10 X0

7
Simulation physics requirements

• Accuracy in the simulation of em interactions,
  down to low energies
• Reasonable simulation of hadronic interactions,
  rather fast
• Plus technical requirements a well written code
• Modularity
• Good documentation
• Maintenability

8
The architecture we want
Geom
Sim data
Phys Sim
Digit
Recon
FAST
Real data
From any point to graphics
9
The beginning GISMO

• The GLAST simulation has been done, from the
  beginning, using C and with OO technologies in
  mind
• GISMO was the choice
• No other candidate present at that moment (apart
  from standard Fortran MC)
• GLAST core software group already experienced
  with GISMO (SLAC used it for other experiments)

10
Characteristics of GISMO

• Takes care of tracking, Eloss etc.
• Secondaty processes EGS4, GHEISHA wrapped in

11
From GISMO to G4

• Why
• GISMO is now quite obsolete
• It is no more officially supported (and
  developed)
• Physics needed some manpower
• GEANT4 has arrived in the meanwhile
• More OO structured (so more flexible,
  maintainable and so on)
• Well supported and used by several experiments
• Continuously developed 2 major releases each
  year monthly internal tag (frequent bug fixes,
  new features, new examples)
• Proved reliable for space applications (XRayTel
  and GammaRayTel)
• Groups involved
• Italy Udine, Trieste, Ferrara, Perugia, Bari (5
  FTE)
• Japan (for the Balloon flight, imminent)

12
Status of the simulation effort

• Present status
• Geometry description persistency with XML
• GEANT4 simulation
• Physics validation ongoing
• Event Display work started
• ROOT prototype, WIRED in evaluation
• Near future
• Integration of the simulation and all the
  software in the GAUDI framework (adopted by
  GLAST)
• Validation of the simulation with beam test and
  balloon data
• Start developing the scientific software

13
The simulation chain (for now)

• Geometry input by XML file
• Incoming fluxes by standard GEANT4 modules
• Simulation with standard GEANT4 physics modules
  (for the em processes also the low energy
  extensions)
• Persistency of the output
• ASCII file
• ROOT file
• Digitization of the MC hits
• Analysis and Event Display
• Validation with real data
• A case study

14
XML for geometry description

• A specific DTD for the GLAST geometry
• A C hierarchy of classes for the XML interface
  (detModel)
• Many clients
• Simulation
• Reconstruction
• Analysis
• Event display
• Interfaces for
• VRML output for the geometry
• HTML documentation
• GEANT4 geometry description
• ROOT
• Java (partial)

15
XML VRML output
16
XML GEANT4 interface
17
Digitizations

• For a more precise digitization of the tracker
  signal
• Electron motion in Si simulation using HEED
  GARFIELD/MAXWELL gt charge sharing (Bari, Pisa)
• Parametrization to be interfaced to the G4
  simulation

18
Event Display

• Various possibilities now in evaluation phase
• WIRED2
• ROOT (directly linked to the G4 simulation
  output)
• Italy (Udine) has this responsibility
• First step User Requirements Document for the
  end of June

19
GEANT4 validation

• Real data
• Experience from developing GammaRayTel, an
  advanced example of G4 (Trieste/Ferrara, Udine)
• Test beam data of 1999 at SLAC (Perugia)
• Balloon data available this summer (Pisa)

20
G4 and GAUDI

• GAUDI has been adopted in the last year from
  GLAST
• It is a framework
• Developed at CERN
• Adopted by various experiments (most important
  LHCb)
• We have to integrate GEANT4 in GAUDI
• Bari is working on that
• LHCb has already done it (Giga)
• We will use it as a starting point

21
A test case study

• To test the simulation chain (simulation,
  reconstruction etc)
• WIMPs annihilation
• Study on Gamma Ray Burst
• The starting point for the next phase
• Developing of the science software (starting from
  next year)

22
Conclusions

• A long way to go
• G4 validation with real data
• GAUDI integration
• Event Display development
• but
• We have proved that G4 is suitable as a MC
  toolkit for space application
• We have already acquired a good experience
• G4 is easy to integrate with other software