Hadronic interaction models and atmospheric showers:

1
Hadronic interaction models and atmospheric
showers the low energy side
Giuseppe Battistoni (INFN Milano) After a talk
given at NEEDS workshop in Karlsrue , April 2002

• Outline
• The case for Microscopic Models
• The experience with FLUKA
• Some basic elements of Microscopic Models of
  FLUKA and DPM Class of simulations
• Accelerator benchmarks
• Applications in cosmic ray physics
• Introduction of the problem

VULCANO Workshop May 2002
2
Calculations for Cosmic Ray Physics requires
reliable hadronic models which can cover a verly
large energy range.
Usually one assumes that the problem exist for
very large energies (extrapolation)
Indeed problem exists also for the "low energy
sector" where "precision models" are requested
from 100 MeV to hundreds of GeVs Atmospheric
neutrinos, muons, hadrons, photons,
... application of particle physics to
astrophysics
The pressure for high quality models comes from

• the experiments (analysis tools)
• now other uncertainties (fluxes
  lt100 GeV) are more under control

• issues coming from Applied Physics
• Example calculation of radiation
  doses for civil aviation

3
The case of atmospheric neutrino is one of the
main examples...
From T.K.Gaisser et al. _at_ICRC 2001 After some
years, the Bartol group realizes that its
orginal neutrino flux calculation is biased by
some inaccuracy of their hadronic model (black
arrow) They propose modification (blue arrow) is
now close to FLUKA (G.B. et al., 1999) results
(red arrow)
Also Japanese group of SK is changing model
switching to DPMJET-III
4
Also in GeV range, main limitations and
uncertainties arise by the lack of a really
calculable theory for the bulk of hadronic
interactions (i.e. far from the perturbative
regime of QCD) Existing experimental data are not
enough to build a global parametrization.
Also, nuclear effects may become very important
to understand how building the passage from
hadron-hadron to hadron-Nucleus Need for
sophisticated nuclear models
In general gt when only phenomenological models
are available, no one of them is good at all
energies and for all cases. Need for combination
of models
This is more problematic at low and intermediate
energies than at UHE, i.e. far from asymptotic
regimes (unless new physics is necessary at
extreme energies we have tools to understand VHE)
5

• In any case, High Energy Codes also need an
  interface
• to a low energy sector
• Hadrons from HE interactions are
• processed in low energy part of the model
• Many showers observables are sensitive to
  low-energy model
• (Example muons at surface level...)
• For instance, CORSIKA makes use
• of GHEISHA below 80 GeV
• There is a deep concern in CORSIKA people
• (and KASKADE group)
• for the unreliable behaviour of this part
• Also some part of low energy codes has to used
• "as it is" also in the High Energy Part
• (example in the following)

6
Calculation Models (MonteCarlo simulations)
Shower (particle transport) model particle
production model
We can think of 2 main classes
"Microscopic" models aiming to describe
interactions at constituent level,
following phenomenological models "theoretically"
inspired
Parametrizations of data, guided by analytical
expressions
Example GHEISHA Bartol code (TARGET) Hillas'
splitting algorithm
Examples FLUKA, UrQMD,DPMJET, QGSJET,
SIBYLL, VENUS...
(linear combinations are possible)
7
Advantages of a Microscopic Model
Allows the application outside the specific
application and range used for tuning it allows
a guided extrapolation

• Very detailed gt aims to precision
• gt necessary for "final" calculations

Good models of this kind have few free
parameters (cannot be handled at your pleasure...)

• On the other hand this means
• not of immediate comprehension,
• difficult to introduce changes

They cannot fit data everywhere with the same
quality
8
The experience with a Microscopic Model the case
of FLUKA
9
The hadronic models of FLUKA

• Main principles which characterize FLUKA
  development
• Quality of algorithms
• High level of accuracy
• (continuous) benchmarking with a wide set of exp.
  Data

10
PEANUT (PreEquilibrium Approach to NUclear
Thermalization) a 3 step model 1. Generalized
Intranuclear Cascade 2. "PreEquilibrium" 3.
Evaporation/Fission or Fermi Break-up (production
of residual nuclear fragments, de-excitation,
etc.)
A demonstration of the importance of nuclear
and quantum effects in particle production at
low-intermediate enegies
Quantum effects Fermi motion, Pauli blocking,
formation zone, etc. particle trajectories
inside nucleus bent by local field
11
(No Transcript)
12
Neutron yield fundamental for dosimetry
problems (example with c.r. civil
aviation...) For neutron treatment below 20
MeV special libraries are used (collaboration
with ENEA/ITALY)
13
The High Energy sector
Phenomenological concepts from Regge-Gribov
theory These inspire also the models like
QGSJET, NEXUS, DPMJET
Color "Chains" (or strings) to be converted in
free particles hadronization
14
That's fundamental the same setting will be
valid at 1011 eV and 1020 eV!! (true for QGSGET,
DPMJET, etc.)
15
(No Transcript)
16
FLUKA BenchMarking

• For recent reviews and extended presentations of
  comparison with data
• (A. Ferrari)
• Proc. 1st International Workshop on Space
  Radiation Research and 11th NASA Space
  Radiation Health Investigators Workshop, Arona,
  Italy, 27 ? 31 maggio 2000
• Proc. MonteCarlo 2000 Lisbona, 2000

For a complete list of references FLUKA server
http//www.fluka.org
17
Benchmarking discussed for atmospheric neutrino
generation
18
(No Transcript)
19
(No Transcript)
20
After 1997 main contribution to hadronization
tuning
Useful for neutrino beam calculations CERN-Gran
Sasso beam engineered with FLUKA
21
Application of FLUKA to the calculation of
secondary particle production by cosmic rays in
the Earth's atmosphere

• Muon fluxes
• Neutrino fluxes
• Hadron fluxes
• Study of transport of primaries in geomagnetic
  field
• (in collaboration with AMS)
• Radiation (neutron) fields in atmosphere

22
Negative muons _at_ floating altitudes CAPRICE94
Open symbols CAPRICE data Full symbols FLUKA
23
CAPRICE DATA FLUKA
24
Horiz. Atmospheric Muon data (DEIS Allkofer et
al.)
25
(3D) Calculation of Atmospheric Neutrino Flux
From a recent review by T.K.Gaisser M.Honda
26
Hadrons in KASKADE calorimeter
Charged hadrons
K0
Neutral hadrons
Histogram FLUKA simulation Circles Exp. data
27
Reproduction of subcutoff structure in primary
protons detected by AMS
28
Calculation of radiation field for risk
evaluation in civil aviation (using FLUKA)
S. Roesler et al. Monte Carlo Calculations of
the Radiation Field at Aircraft Altitudes,
SLAC-PUB-8968, to be published in Radiation
Protection Dosimetry (2002)
ambient dose eq. From neutrons at solar maximum
on commercial flights from Seatte to Hamburf
and from Frankfurt to Johannesburg
29
A collaboration with CORSIKA group is under
way to insert FLUKA as low-energy part of the code

• Other applications connected to CR physics, but
  not to
• atmospheric showers
• Radiation dose studies for long duration space
  missions
• (collaboration with NASA and Houston Univ.)
• Radiation damage in spacecrafts

Accelerator Physics FLUKA is part of the
present baseline for simulation in ALICE _at_
LHC Also ICARUS
30

• Below the knee region, CR physics is entering a
  "precision" era,
• where also precision tools are needed

• The "microscopic" approach to hadronic
  interactions
• in our opinion can guarantee reliability in a
  wide range of energies.

• FLUKA and similar models seem to give optimal
  results
• both in accelerator physics and c.r. physics

• The same kind of models gives the most reliable
  (at present!)
• guideline for extrapolation to UHE

• The accuracy in calculation models applied to
  cosmic ray physics
• is fundamental for the reliability of exp. data
  analysis and
• also for applications outside fundamenta research!

• Of course no MC model is perfect (since we have
  no really calculable theory) a lot of work is
  still necessary for improvement (exp. Data
  modelling)

31
from h-h to h-N (and to N-N) cross sections
From total and elastic cross sections (PDG)
isospin decomposition Glauber model exp.
Data
32
Shortcomings of the parametrized approach the
case of GHEISHA
Tests performed for calorimetric applications in
LHC exp. (0.5 - 100 GeV)
ALICE-INT-2001-41 (2001) ATLAS PHYS-86 (1996)
Failures in conservation laws Non uniformity in
azimuthal distribution In general GHEISHA was
born at the time of early exp. _at_DESY gt
parametrizations are always dangerous
outside their validity range
Similar problems appeared in the framework of
neutrino beam calculations
33
(No Transcript)
34
(No Transcript)
35
FLUKA
GHEISHA (GEANT-3)