Cosmic rays rays at at PeV PeV energies energies with with - - PowerPoint PPT Presentation
Cosmic rays rays at at PeV PeV energies energies with with Cosmic GAMMA Experiment Experiment GAMMA Anatoly Erlykin on behalf of the GAMMA collaboration Vulcano Workshop May 2010 ARAGATS scientific station (late autumn) Hill sides
Vulcano Workshop May 2010
Anatoly Erlykin
ARAGATS scientific station (late autumn) Hill sides of the Mt. Aragats, Armenia, 65 km from Yerevan Elevation: 3200 m a.s.l. (700 g/cm2 of atmospheric depth) Geographical coordinates: Latitude = 40.470 N, Longitude = 44.180 E
Construction of the GAMMA was begun in the middle of 80th in the frame of joint project ANI of the Moscow Lebedev Institute and Yerevan Physics Institute. Ambitious project for studying CR at 0.1 – 1000 PeV including
(half-built)
big electromagnet, weight 3000 ton (half-built) After the collapse of the Soviet Union in the beginning of 90th construction was practically stopped for about 5 years. Nevertheless in the end of 90th we succeed to create (in the framework of ANI project) the GAMMA installation with Surface part of scintillation detectors Carpet of underground muon scintillation detectors Start for operation - in 1998
at 1014 -1017 eV Main topics (most of them in progress)
Study of EAS characteristics
at 1014 -1017 eV Main topics (most of them in progress)
Study of EAS characteristics
Particle physics
at 1014 -1017 eV Main topics (most of them in progress)
Study of EAS characteristics
Particle physics
Astrophysical problems, gamma-astronomy
radiation at energies 0.1-100 PeV
GAMMA at present (after several modifications) Surface part (electromagnetic component)
scintillation detectors (S=1m2) in each station. Total number – 116 The area – ~ 30.000 m2
characteristics
Underground part (muon component)
and energy threshold Eµ > 5 GeV)
0.09m2
[Astroparticle Physiscs, 28 (2007) 169] On the base of EAS data the energy spectra and elemental composition of the PCR are derived in the 1 – 100 PeV. The reconstruction of spectra carried out using an EAS inverse approach in the frameworks of the SIBYLL2.1 and QGSJET0.1 interaction models and the hypothesis of power-law primary energy spectra with rigidity-dependent knees.
The measured variable distributions in comparison with expected dependences from SIBYLL and QGSJET interaction models
m Ri 50 <
Detected and expected particle density spectra measured by surface and underground scintillators for different shower sizes Good agreement besides the muon density for shower size Nch > 107
EAS size spectra for three zenith angle intervals EAS truncated muon size spectra for three zenith angle intervals
EAS size spectra for different truncated muon size thresholds EAS truncated muon size spectra for different shower size thresholds
Dependence of the average EAS age parameter on EAS size Average EAS truncated muon size versus EAS size
Dependence of the average EAS age parameter on EAS size Average EAS truncated muon size versus EAS size
B.Wiebel & P.Biermann, 24th ICRC (1995) A.Lagutin et al., 29th ICRC (2005)
The reconstruction of the primary energy spectra is carried out using an inverse approach for simulated data base (SIBYLL2.1 and QGSJET01 interaction models and the hypothesis of power-law primary energy spectra with rigidity-dependent knees.
Extrapolation of balloon and satellite data to ~ 103TeV
up to E~100 PeV.
p, He, O-like and Fe-like nuclei strongly depend on interaction model.
consistency of the extrapolation of derived primary spectra with direct measurements in the energy range of satellite and balloon experiments.
weakly depend on interaction model.
parameter for EAS size Nch > 107 is observed and requests additional analysis.
[J.Phys. G: Nucl. Part. Phys. 35 (2008) 115201]
On the basis of extended EAS data set from the GAMMA experiment an all-particle primary CR energy spectrum in the 3- 200 PeV energy range was obtained by a multi-parametric event-by-event evaluation of the primary energy. The energy evaluation method has been developed using the EAS simulation with the SIBYLL interaction model taking into account the response of the GAMMA detectors and reconstruction uncertainties of EAS parameters.
where Nch, Nµ, s, cos θ – experimentally measured parameters
The best energy estimations as a result of χ2
min(E0,E1) were
achieved for the 7-parametric fit:
where x = LnNch, y = LnNµ (R<50m), c = cos(θ)
b ≈ +0.10 and b ≈ -0.17 δ ≈ 1 ± Δδ(Ε)
Mean biases versus energies of the primary proton (p) and iron (Fe) nuclei and the uniformly mixed p, He, O, Fe composition (All) The boundary lines corresponds to approximations with upper and low limits The shaded area corresponds to approximations
b ≈ +0.09 and b ≈ -0.15
and were used to estimate errors for reconstruction of the all- particle energy spectrum
σ = 0.14, Δσ = 0.03 σA(E) ≈ σ(E)
Errors of the energy estimator versus primary energy E0 for 4 primary nuclei and uniformly mixed (All) composition. The cross symbols are taken from our Previous data computed for the mixed composition and shower core selection criteria R<25m
!!! Such high accuracies of the energy evaluation regardless of primary nuclei
are a consequence of the high mountain location of the GAMMA facility (700 g/cm2), where the correlation of primary energy with the detected EAS size is very high (about 0.95) !!!
E0-E1 scatter plots of simulated primary energy E0 and estimated energy E1(Nch,Nµ,s,θ) for 4 primary nuclei
R<50m, θ<450
Detected zenith angular distributions for different energy thresholds. The lines correspond to simulated distributions with the same statistics. The agreement of detected and simulated distributions gives an additional support to the consistency of energy estimates in the whole measurement range. The anisotropic spectral behavior at low energy (less than 3 PeV) is explained by the lack of heavy nuclei at larger zenith angles.
24
GAMMA05: R < 25m; Q < 300 GAMMA07: R < 50m; Q < 450 All-particle energy spectrum in comparison with the results of EAS inverse approach (GAMMA-06, KASCADE, KASCADE-Grande), our preliminary data and results of
Dependence of the average EAS age parameter on EAS size Average EAS truncated muon size versus EAS size
Rigidity-dependent primary-energy spectra cannot describe the phenomenon of ageing of EAS at energies (5-10) x 1016 eV which was
It is reasonable to assume that an additional flux of heavy nuclei (Fe- like) is responsible for the bump at these energies. Besides, the sharpness of the bump points out the local origin of this flux from compact
We carried out the test of this hypothesis using the inverse approach on the base of GAMMA data and the hypothesis of two- component origin of cosmic ray flux: so-called Galactic component is the power-law energy spectra with rigidity-dependent knees at energies Ek=ER·Z and power indices γ = γ1 and γ= γ2 for E < Ek and E > Ek respectively; so-called pulsar component is an additional power-law energy spectrum with cut-off energies Ec,Fe and indices γp= γ1,p and γp= γ2,p for E < Ec,Fe and E > Ec,Fe respectively.
29
EAS size (on the left) and truncated muon size spectra and corresponding expected spectra computed in the framework of the SIBYLL interaction model and the two-component parametrization of primary spectra. The lines correspond to expected spectra computed for each of the primary nuclei.
30
All-particle energy spectrum and expected energy spectra derived from EAS inverse problem solution for p, He, O and Fe primary nuclei using two-component parametrization along with
Galactic p, He, O and Fe components All-particle energy spectrum and expected energy spectra derived from EAS inverse problem solution for p, He, O and Fe primary nuclei using two-component parametrization along with energy spectra of Galactic p, He, O and Fe components and with additional Fe component from compact objects
All-particle energy spectrum and expected energy spectra derived from EAS inverse problem solution for p, He, O and Fe primary nuclei using two-component parametrization along with energy spectra of Galactic p, He, O and Fe components and with additional Fe component from compact objects !!! Two-component all-particle spectrum (bold line with shaded area) agree within the errors with the results of the event-by-event analysis !!! Galactic p, He, O and Fe components Fe component from compact objects
ll-particle energy spectrum are obtained using GAMMA facility EAS database. The all-particle spectrum in the range of statistical and methodical errors agrees with the same spectrum
High accuracies of energy evaluations and small statistical errors point out to the existence of explicit irregularity (bump) of energy spectrum in the 60-80 PeV region.
provided high accuracies (10-15%) for energy evaluation of primary nuclei regardless of nuclei mass number in 5-200 PeV energy region; he bump can be described by a two-component model of primary CR origin with additional (pulsar) Fe components. hough we cannot reject the stochastic nature of the bump completely, our examination of the systematic uncertainties of the applied method lets us believe that they cannot be responsible for the observed feature.
!!! It is necessary to increase statistics and to verify again all-particle energy spectrum !!!
Confirmation of GAMMA results by other experiments
Confirmation of GAMMA results by other experiments ‘Iron’ knee ?
[31st ICRC, Lodz, 2009] An upper limit of galactic diffuse gamma-ray flux, measured with the GAMMA experiment at energy about 175 TeV. The results were obtained using selection of muon poor extensive air showers for 5 GeV muon energy threshold.
Discrimination of the γ-showers from primary induced showers is performed
1) the reconstructed shower core coordinates is distributed within radius of R < 15m; 2) shower zenith angles θ < 300; 3) reconstructed shower size Nch > 105; 4) reconstructed shower age parameters (s) is distributed within 0.4 < s < 1.5; 5) goodness-of-fit test for reconstructed showers χ2 < 2.5; 6) no-muon signal is recorded for detected showers satisfying the previous 5 conditions. The selection criteria and γ-shower discrimination rule were obtained using CORSIKA shower simulation code for the NKG and EGS modes in the frameworks of the SIBYLL interaction model. Simulation were done for 4 nuclear species: p, He, O and Fe.
Upper limit of gamma ray flux derived from detected no-muon showers (black triangle symbol). The gray symbols are the CASA-MIA [11], KASCADE [13] and EAS-TOP [12] data taken from [13]. The lines are expected Galactic diffuse background flux from [Aharonian, F. A., A. M. Atoyan 2000, Astron.
An upper limit of γ-ray differential flux at energy Eγ ~ 175 (+25, -20) GeV obtained with GAMMA experiment is equal to (5.8 – 7.0) · 10-12 (erg · m2 · s · sr)-1 for 95% confidence level and it is in close agreement with the CASA- MIA data [11]. The lower limit for the primary energy spectra and elemental composition obtained with the GAMMA experiment [14] can be extended to the lower energy region up to about 100 TeV energies.
I Increasing statistic for more detailed study of structure of the primary cosmic ray energy spectrum around the knee at energies 1 – 100 PeV
The development of modern gamma-astronomy and particularly the study of the characteristics of the diffuse γ-ray flux at energies greater than 1014 eV are connected, practically without any alternative, with the registration and the selection of EAS with an abnormally small relative content of muons and hadrons.
detectors mounted in the central part of the GAMMA array will allow
250 sq.m using Geiger counters for a selection EAS generated by gamma-rays.
ARMENIA Yerevan Physics Institute R.Martirosov, A.Garyaka RUSSIA Moscow Lebedev Institute A.Erlykin, N.Nikolskaya USA L.Jones, University of Michigan S.Ter-Antonyan, Southern University, Baton Rouge France Y.Gallant, University of Montpellier J.Procureur, University of Bordeaux
COLLABORATION
Trigger efficiency of GAMMA EAS array for different primary particles
Expected reconstruction error (upper panel) and average bias (lower panel) of shower size (Nch) for different primary particles (symbols). Δch=ln(Nch
*/Nch) and Nch * is an estimation of Nch
Shower age parameter (s) distribution for all showers (left panel) and no-muon detected showers (N¹ = 0, right panel). Simulated data for the primary mixed composition (All nuclei, left panel) and primary gamma ray (right panel) are normalized to the corresponding GAMMA experimental data.
Primary energy spectra and all-particle energy spectrum taken from Astropart. Phys. 2007 (shaded area) and corresponding extrapolations to the 100 TeV energy region.
Normalized detected muon number (Nµ) spectra for different shower size thresholds (105; 2 x 105; 4 x 105). Hollow symbols (circle, square and triangle) are GAMMA experimental data. The symbols in γEGS, (p - Fe)EGS and (p - Fe)NKG columns correspond to simulated data for the primary γ and mixed composition (p;He;O; Fe) computed using the EGS and NKG modes of CORSIKA.
Primary energy (E) and corresponding shower size (Nch) distributions at
dot symbols). Hollow circles correspond to the proton showers with no-muon signal from underground muon carpet (Nµ = 0). Solid and dashed lines are the log- linear approximations (see text) for primary γ and p correspondingly. Inset histograms are E p,γ /Nch distributions for primary proton and γ (shaded area).
Detected (histogram lines) and expected (symbols) muon number (Nµ) spectra for different shower size thresholds (105; 2 x 105; 4 x 105) and different mode (NKG, EGS) of CORSIKA