Test of the SIBYLL 2.3 high-energy hadronic interaction model using the KASCADE-Grande
muon data
1Tests of SIBYLL 2.3 using KG data - J.C. Arteaga ISMD 2017, Tlaxcala, Mexico
Juan Carlos Arteaga-Velázquez*, D. Rivera for the KASCADE-Grande CollaborationInstituto de Física y Matemáticas, Universidad Michoacana, México
1. Introduction
2. Motivation
3. The KASCADE-Grande detector
4. Data & Simulations
5. Analysis
6. Results
7. Summary
Outline
Test of the SIBYLL 2.3 high-energy hadronic interaction model using the KASCADE-Grande
muon data
Juan Carlos Arteaga-Velázquez*, D. Rivera for the KASCADE-Grande CollaborationInstituto de Física y Matemáticas, Universidad Michoacana, México
2ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
30 - 20 km
Cosmic rays are produced in HE astrophysical sources (SNR’s, AGN’s, etc?).
Primaries with E > 1 PeV are detected at Earth through air shower (EAS) observation
EAS data is interpreted with hadronic models to study energy and composition
Introduction
3ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Hadronic interaction models:1. Phenomenological models inspired in QCD.
3. Calibrated with accelerator data.
4. Extrapolated to high energies (HE’s) and forward region (pT ~ 0).
Soft physics (low Q2) is relevant for CR interactions
Model uncertainties produce uncertainties in predictions of EAS parameters
Introduction
T. Pierog,EPJ web of conferences 145, 18002 (2017) 0.1 % 0.02 % 99.88 %
4ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
T. Pierog,EPJ web of conferences 145, 18002 (2017)
Differences in EAS observables due to uncertainties in the models
Introduction
5ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Composition and energy scale a r e a f f e c t e d b y m o d e l uncertainties
M. Bertaina et al., Pos(ICRC2015) 359
KASCADE Coll., Astrop. Phys. 24 (2005) 1.
All-particle spectrum
Dependence of relative abundances and spectrum of CR’s with hadronic interaction models:
Mass groups Mass groups
Imperative to check validity of hadronic models
KASCADE-Grande experiment
Introduction
6ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Proton @ 1015 eV, Corsika simulation, F. Schmidt & J. Knapp
Employ muons for tests:
- Penetrating particles/less atmospheric attenuation.
- Sensitive to hadronic processes.
- Used in composition studies.
Use CR observatories to constrain/test models:
- KASCADE-Grande
- Pierre Auger
- EAS-MSU
ECR = 1015 - 1018 eV Ethμ = 230 MeV, 490 MeV, 800 MeV, 2.4 GeV
ECR = 1015 - 1017 eV Ethμ = 0.2 GeV
ECR > 1018 eV Ethμ = 1 GeV
ECR = 1017 - 1018 eV Ethμ = 10 GeV
- ICECUBE/ICETOP
- Keep information from early stage of EAS development.
Introduction
7ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Proton @ 1015 eV, Corsika simulation, F. Schmidt & J. Knapp
Muon measurements:
Use CR observatories to constrain/test models:
- KASCADE-Grande
- Pierre Auger
- EAS-MSU
ECR = 1015 - 1018 eV Ethμ = 230 MeV, 490 MeV, 800 MeV, 2.4 GeV
ECR = 1015 - 1017 eV Ethμ = 0.2 GeV
ECR > 1018 eV Ethμ = 1 GeV
ECR = 1017 - 1018 eV Ethμ = 10 GeV
- ICECUBE/ICETOP
Introduction
- Energy spectrum
- µ-/µ+Charge ratio
- Multiplicity
- Zenith angle dependence
- Lateral distributions
- Production height
- Pseudorapidites
8ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
KASCADE-Grande EAS muon data
Muon Eth > 230 MeV x Secθ
Muon attenuation length (Λµ):
1. Parameterizes dependence of number of μ’s in EAS with the atmospheric depth:
2. Correct data for attenuation in the atmosphere.
3. Affected by details of shower production:- π energy spectrum,
cross section, pT distribution,
- π±/π0 ratios,
- Baryon/resonance production,
- Multiplicity <N>
- Inelasticity (y), etc.
Motivation
9ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Nμ = Nμo e-(X/Λμ)
5 5.5 6 6.5 7
)2 (g
/cm
µΛ400
600
800
1000
1200
1400
1600 )° - 40° = 0θKG data (
QGSJET II-2 EPOS LHCSIBYLL 2.1QGSJET II-04
Measured muon attenuation length (ECR ~ 1016 - 1017 eV) is above MC predictions from:
J.C. Arteaga et al., Astropar. Phys. 95 (2017) 25
KASCADE-Grande EAS muon data
Does SIBYLL 2.3* perform better?
Pre-LHC models (~ 2 σ):- SIBYLL 2.1 - QGSJET-II-02
Post-LHC models (~ 1.34 σ to 1.48 σ):- EPOS-LHC - QGSJET-II-04
Better agreement with post-LHC models.
Motivation
*F. Riehn et al., PoS(ICRC2015) 558
[ ] Total unc. — Stat. unc.
10ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Less effective attenuation in exp. data
Muon attenuation length (Λµ):
Data
-Spectrum -Composition -Arrival direction
knee
2nd knee
γ = -2.7γ -3.0
γ ~ -3.3
γ = -2.6
J(E) = E-γ
The KASCADE-Grande detector
11ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Data
-Spectrum -Composition -Arrival direction
knee
2nd knee
γ = -2.7γ -3.0
γ ~ -3.3
γ = -2.6
J(E) = E-γ1. What is the origin of the features in the spectrum?
2. Where do they come from?
3. What is their nature?
4. How do they get accelerated?
5. Are there nearby sources?
6. Where is the galactic t o e x t r a g a l a c t i c transition?
The KASCADE-Grande detector
12ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
1. Location: KIT-Campus North, Karlsruhe, Germany
December 2003 - November 2012
Karlsruhe, Germany110 m a.s.l., 49o N, 8o E
The KASCADE-Grande detector
13ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
Grande array KASCADE array
W.D. Apel et al., NIMA 620 (2010) 490
37 plastic scintillator detectors
137 m
E= 1 PeV - 1018 eVKASCADE (200 x 200 m2) + Grande (0.5 km2)
252 shielded/unshielded scintillator detectors, muon tunnel, calorimeter.
The KASCADE-Grande detector
14ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
KASCADE array
W.D. Apel et al., NIMA 620 (2010) 490
Scintillator detectors
• Charged particles (> 3 MeV)
H. Falcke et al., Nature 435 (2005) 313
• Ne (> 5 MeV) and Nµ (> 230 MeV)
e/γ detector
µ detectorPb/Fe shielding
liquid scintillator
plastic scint.
Grande array
KASCADE (200 x 200 m2) + Grande (0.5 km2) E= 1 PeV - 1018 eVThe KASCADE-Grande detector
15ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
1. Grande provides Nch: Number of charged particles
2. KASCADE provides Nµ : Number of muons
ρµ(r) = Nµ ⋅ fµLagutin(r)Fit to data:
Fit to data:ρch(r) = Nch ⋅ fchNKG(s, r)
The KASCADE-Grande detector
16Tests of SIBYLL 2.3 using KG data - J.C. Arteaga
The KASCADE-Grande detector
Unfolding methods capable of reconstructing spectra of elemental groups:
- Knee at E~1015 eV due to a break in the spectrum of light components
- Spectral features independent of the hadronic interaction models
- Iron knee around 80 PeV
Exploit Ne-Nµ correlation Exploit Nch-Nµ correlation
Knee positions ∝ Z
QGSJET-II-02/FlukaKASCADE Coll., Astropart.
Phys. 24 (2005) 1
KASCADE'Grande.Coll.,.Astropart..Phys..47.(2013)
17ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
KASCADE Coll., Astrop. Phys. 36 (2012) 183
- Knee structure around 80 PeV in the heavy component
- Ankle-like feature at 120 PeV in the light component
Heavy
Light
Galactic-extragalactic transition?
2nd knee
2nd knee
QGSJET-II-02
18ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga
The KASCADE-Grande detector
1. Effective time: 1434 days
2. Área: 8 x 104 m2
3. Exposure: 2.6 x 1012 m2 s sr
4. Cuts (reduction of EAS uncertainties): - Central area - θ < 40o - Instrumental & reconstruction cuts - Optimized for E = [1016, 1017] eV
Experimental data
Data & simulations
2 744 950 selected events
Efficiency: log10 (E/GeV) = 7 ± 0.20 log10 (Nμ) = 5 ± 0.20
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 19
1. HE hadronic interaction Model:
2. Simulation: H, He, C, Si, Fe, mixed;
3. Systematics: - ΔΝch < 12% ΔΝμ < 20% - Δθ < 0.6o - σcore < 10 m
Νμ data corrected for systematic errors
MC data (CORSIKA/Fluka)
SIBYLL 2.3
γ = -3, -3.2, -2.8
ΔΝμcorrected < 10%
Data & simulations
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 20
θ < 42o
E = 1014 - 3 x 1018 eV
)θsec(1 1.05 1.1 1.15 1.2 1.25 1.3
)µ
(N10
log
5.15
5.2
5.25
5.3
5.35
5.4
5.45 MC data (Mixed)X 0
X0 sec(θ)
E E
Nμ(θ)Nμ(0)
θ
Ε ~1016 eV
SIBYLL 2.3
Shower content at same Energy (E) is attenuated with atmospheric depth (X):Large X —> High zenith angles (θ)
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 21
Analysis
)θsec(1 1.05 1.1 1.15 1.2 1.25 1.3
)µ
(N10
log
5.15
5.2
5.25
5.3
5.35
5.4
5.45 MC data (Mixed)X 0
X0 sec(θ)
J(E) J(E)
J[Nμ(θ)]
J[Nμ(0)]
θ
Ε ~1016 eV
SIBYLL 2.3
J = constant
Constant intensity cut method100 % detector efficiency + Isotropy —> J[Nμ(0)] = J[Nμ(θ)]
- Constant Intensity Cut method: Quantify zenith-angle evolution of data.
- Method is independent of MC model.
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 22
Analysis
Data divided in five θ intervals with equal exposure.
)µ
(N10
log5 5.5 6 6.5 7
)-1
sr
-1 s
-2) (
mµ
J(>N
-1310
-1210
-1110
-1010
-910
-810 KG data
o < 16.71θ ≤ o0.00o < 23.99θ ≤ o16.71o < 29.86θ ≤ o23.99o < 35.09θ ≤ o29.86o < 40.00θ ≤ o35.09
)θsec(1 1.05 1.1 1.15 1.2 1.25 1.3)µ
(N10
log
4.8
5
5.2
5.4
5.6
5.8
6
6.2
6.4
[J] = -9.8010
log
[J] = -8.6010
log
KG Data
1. Apply cuts at fixed frequencies 2. Get attenuation curves
3. Apply a fit to get Λμ
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 23
Analysis
J.C. Arteaga et al., Astropar. Phys. 95 (2017) 25
Results
ΔΛμ QGSJET-II-2 QGSJET-II-4 SIBYLL 2.1 SIBYLL 2.3 EPOS-LHC
σ +2.04 +1.48 +1.99 +1.06 +1.34
5 5.5 6 6.5 7
)2 (g
/cm
µΛ
400
600
800
1000
1200
1400
1600 )° - 40° = 0θKG data (
QGSJET II-2 SIBYLL 2.3SIBYLL 2.1QGSJET II-04 EPOS LHC
MC data also include: - Errors from composition - Unc. from spectral index of CR
intensity
[ ] Total unc. — Stat. unc.
*Errors on SIBYLL 2.3 are preliminary
Discrepancy between SIBYLL 2.3 and measurement is small, but large uncertainty from composition
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 24
MC data points: Mixed composition
Results
Reduce error due to composition uncertainties:
- Χ2 fit to measured data with 4 mass groups: H, He, C, Si+Fe (50 % mixture).
- Use double power-law for energy spectrum of each mass group.
- Employ templates from SIBYLL 2.3 for each mass group.
)µ
(N10
log4 4.5 5 5.5 6 6.5 7 7.5
)ch
(N10
)/log
µ(N
10lo
g
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
Events
1
10
210
310
410
510
KG data° - 40° = 0θ
12.392 million events
A: atomic mass
J.C. Arteaga et al., (KG Collab.) PoS (ICRC2017) 316
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 25
Full data
Energy (eV/particle)1610 1710 1810
)1.
5 e
V-1
sr
-1 s
ec-2
J(E
) (m
2.5
dI/d
E x
E
1410
1510
1610
1710
1810
1910
KG (SIBYLL 2.3)HHe+CHeavy
Akeno (J.Phys.G18(1992)423)
AGASA (ICRC 2003)
HiResI (PRL100(2008)101101)
HiResII (PRL100(2008)101101)
Yakutsk (NewJ.Phys11(2008)065008)
AUGER (ICRC 2013)
EAS-TOP (Astrop.Phys.10(1999)1)
TIBET-III (ApJ678(2008)1165)
GAMMA (J.Phys.G35(2008)115201)
TUNKA (Nucl.Phys.B,Proc.Sup.165(2007)74)
KASCADE (QGSJET01 Astrop.Phys.24(2005)1)
KASCADE (SIBYLL2.1 Astrop.Phys.24(2005)1)
µKASCADE-Grande (QGSJET II) Nch-N
µKASCADE-Grande (EPOS-LHC) Nch-N
KG (SIBYLL 2.3)
Results
Composition model obtained from measured data using SIBYLL 2.3
J.C. Arteaga et al., (KG Collab.) PoS (ICRC2017) 316
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 26
5 5.5 6 6.5 7
)2 (g
/cm
µΛ
400
600
800
1000
1200
1400
1600 )° - 40° = 0θKG data (
QGSJET II-2 SIBYLL 2.3SIBYLL 2.1QGSJET II-04 EPOS LHC
Results
ΔΛμ QGSJET-II-2 QGSJET-II-4 SIBYLL 2.1 SIBYLL 2.3 SIBYLL 2.3 EPOS-LHC
Composition model
σ +2.04 +1.48 +1.99 + 1.06 + 1.52 +1.34
[ ] Total unc. — Stat. unc.
*Errors on SIBYLL 2.3 are preliminary
SIBYLL 2.3 has also problems to describe the data
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 27
Results
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 28
J.C. Arteaga & D. Rivera et al., (KG Collab.) PoS (ICRC2017) 316
Muon lateral densities]
-2(r)
[mµρ
1−10
1
10
protonsFeKG data
Sibyll 2.3 ]o,16o [0∈ θ
[6.55,6.8]∈ chN10
log
r [m]200 400
]-2
(r) [m
µρ
1−10
1
10 ]o,35o [30∈ θ
[7.04,7.28]∈ chN10
log
r [m]200 400
[7.52,7.74]∈ chN10
log
r [m]200 400
Energy
Zenith
chN10
log6 7
µN10
log
4.5
5
5.5
6
6.5
7
protonsFeKG data
Sibyll 2.3 ]o,16o [0∈ θ
chN10
log6 7
]o,30o [24∈ θ
chN10
log6 7
]o,40o [35∈ θ
Results
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 29
Nµ - Νch correlation
Zenith
J.C. Arteaga & D. Rivera et al., (KG Collab.) PoS (ICRC2017) 316
1. The measured Λµ at KASCADE-Grande is above predictions of HE hadronic interaction models: QGSJET-II-02, QGSJET-II-04, EPOS-LHC and SIBYLL 2.3.
2. Post-LHC models predict a Λµ value higher than that predicted by Pre-LHC models.
3. The models might need:
- a harder μ energy spectrum,
- a decrease of elasticity in pion interactions,
- a reduction of forward production of baryon/antibaryon pairs, etc.,
to agree with the data.
Summary
ISMD 2017, Tlaxcala, MexicoTests of SIBYLL 2.3 using KG data - J.C. Arteaga 30
Thank you!