CORSIKA/MMC for the IceCube simulation production

1
CORSIKA/MMC for the IceCube simulation production
Dmitry Chirkin
chirkin_at_physics.berkeley.edu University of
California at Berkeley, Berkeley, CA 94720,
USA IceCube Detector Simulation workshop,
Belgium, October 2003
2
Talk outline

• generators/propagators what are they good for?
• description of optimizations done for AMANDA-II
• CORSIKA
• MMC
• conclusions

3
Simulation chain as used in AMANDA

• known muon propagators
• MMC
• MUM
• mudedx
• Lip
• Loh

generator step

• known air shower
• generators
• CORSIKA based
• dCORSIKA
• pCORSIKA
• basiev

propagator step

• Also new possible applications
• CORSIKA gt atm. ? generation
• Kosta Schinarakis, Wuppertal
• MMC gt monopole propagation
• Peter Niessen, Arvid Pohl

4
CORSIKA settings

• 6 interaction models, QGSJET, VENUS and NEXUS
  describe AMANDA-II data best, QGSJET is by far
  the fastest
• Atmosphere and Earths surface are curved
• Detector is assumed cylindrical

5
Running CORSIKA consists of 2 steps

• CORSIKA (shower generation)
• UCR (Utility for CORSIKA output Randomization
  shower location randomizer)

6
Running MMC

• Basic features
• can propagate muons with fixed vcut or ecut in
  regions 1
• and 2 (use negative values to change default
  behavior)
• propagates muons with up to 1020 eV, can be used
  up to
• 1030 eV (enable with ebig1.e21 flag)
• can propagate muons and/or taus. To propagate
  both,
• run MMC for taus first, then pipe into MMC for
  muons
• Can propagate muons though multiple media with
• spherical or flat interfaces. For concentric
  media uses
• external file with media location definitions
• User line can be enabled for each muon. To
  ensure
• compatibility with other AMANDA software, use
• -rdmc (for f2k) and/or -amasim (for AMASIM)
• provides multiple output formats (3 at the
  moment)

Vcut0.05, 0.01, 10-3 and 10-4 Usual (no cont)
vs. cont option

• Defalt values used for AMANDA
• vcut0.05, cont
• ecut500 MeV no cont
• vcut1.0 no cont

User line definition USER_DEF mmc_en NR E_INI
E_CPD E_IN E_OUT CDP_X CDP_Y CDP_Z Z_IN Z_OUT
7
MMC optimizations cont.

• Interpolation routines and strategy of their use
  have been heavily optimized
• Parameterization and interpolation settings can
  be easily reconfigured (e.g. with -gromberg
  option at run time or with -hifi option during
  compilation)
• Algorithm works to the extreme tested up to the
  muon energy of 1030 eV and vcut as low as 10-15.

8
Conclusions and dicussion

• CORSIKA and MMC have been heavily optimized
• Please note that some optimization settings are
  AMANDA-II specific and need to be recalculated
  for desired IceCube geometry