UrQMD and its Application to Cosmic Rays

1
UrQMD and its Application toCosmic Rays

• Marcus Bleicher
• Institut für Theoretische Physik
• Goethe Universität Frankfurt
• Germany

In collaboration with H.J. Drescher
2
Thanks to

• Hajo Drescher
• Sascha Vogel
• Xianglei Zhu
• Stephane Haussler
• Hannah Petersen
• Diana Schumacher

3
Some history

• IQMD, QMD non-relativistic models (Elt 2 AGeV)
• RQMD (first relativistic transport model),
  developed in Frankfurt. Not supported since
  2000
• UrQMD (development started 1996 at Frankfurt)

• NOT transport/cascade models
• HIJING
• PYTHIA/FRITIOF
• NEXUS, VENUS
• DPM

4
The tool UrQMDv2.2

• Non-equilibrium transport model
• Hadrons and resonances
• String excitation and fragmentation
• Cross sections are parameterized
• via AQM or calculated by detailed balance
• pQCD hard scattering at high energies (not in
  v1.3)
• Generates full space-time dynamics of hadrons
  and strings

5
Included Particles
6
Reaction stages

• Initialization of projectile and target
  (Lorentz contracted Woods-Saxon)
• Generate table with collision/decay sequence
  with
• Propagate to next collision
• Perform collision according to cross sections -
  elastic scattering - inelastic scattering
  - resonance production - soft string
  formation and fragmentation - pQCD hard
  scattering / fragmentation
• Update particle arrays, update collision table,
  perform next collisions

7
Resonance cross sections
8
Collision Spectrum

• Initial baryon-baryon scatterings
• Cooking of QCD matter (Thermalization)
• Freeze-out stage

9
Basic checks (I)
10
Basic Checks (II)
Unfortunately the data has poor qualityOne has
to rely on the extrapolation This leads to 10
systematic uncertainty
E21000 AGeV
11
Baryon Stopping
AuAu
AuAU
160 AGeV
21 ATeV
Energy deposition is OK
Anything special here?
12
Pion induced reactions

• UrQMD works reasonable

13
Proton induced reactions
Particle production in pA is in line with data
in UrQMD
14
Particle Production at higher energies
15
Stopping at high energies
16
Ratio of LDFs
p _at_ 1019 eV
LDFs at large distances depend strongly on the
low energy model
17
When to switch from high to low energy model
Corsika switches models at E80 GeV However,
even at 200 GeV, SIBYLL and QGSJET
under/overestimate the pA data
18
Energy dependence

• UrQMD and SIBYLL predict similar results up
  to E1000 GeV
• UrQMD is tested up to 21 ATeV beam energy
• Transition to QGSJET/SIBYLL might be better
  at higher energies (E1000 GeV)

19
Why use UrQMD?

• It is distributed with Corsika
• Its free software i.e. ? Check how we did it
  ? Compile it with your favorite options
  on you favorite system
• It is checked with accelerator data
• It works very well

download from http//www.th.physik.uni-frankfurt.d
e/urqmd