1. Size and Energy Spectra of incident cosmic radiation obtained by the MAKET - ANI surface array on mountain Aragats. (Final results from MAKET-ANI detector)‏ A.Chilingarian, G.Hovsepyan Cosmic Ray Division, Alikhanyan Physics Institute, Armenia.

2. MAKET-ANI detector d esigned to measure Extensive Air Showers (EAS) from Primary Cosmic rays (PC) with energies 10 5 – 10 8 GeV. Detector position: 3200m above see level on Mt. Aragats, Armenia. Geographic location: 40.5 o N, 44.2 o E

3. Consist – 92 plastic scintillation detector 68 with 1m 2 area, and 24 – 0.09 m 2. 19 of 1m 2 area detectors for fast timing system. Two types triggers are used: »for density detectors 7 from 11 »for fast timing system 4 from 9 –Effective area for EAS collection ~900m 2 for N e >10 5

5. is the normalization factor (Hayakawa, 1969, Aseykin, 1979 )‏ The accuracy of the reconstructed EAS core position within the collected area S eff =880m 2 is less than 1 m The reconstructed shower size errors (including systematic) are less than ~13% at Ne =10 5 and quickly decrease with the rise of shower size. The errors of reconstructed age para- meter are less than ~9% at Ne =10 5. angular accuracy not worse than ~1.5 0 for zenith θ, and for the azimuth ones Δφ ~ 5 0 for the showers within 15 0 < θ < 45 0. event { N e, S, X 0,Y 0, Θ, φ} r m =118 m

6. From 1997 to 2004 ~1.2 ∙ 10 7 shower triggers were registered. The total exposition time was ~1.42∙10 8 s. Installation Stability of the operation time. additional restrictions on the EAS triggers N e ≥ 10 5, 0.3 < s < 1.7, θ ≤ 46.8 o, {X 0, Y 0 }  S eff reduce the total number of events by an order of magnitude to a value of ~ 1.3 ∙ 10 6.

7. MAKET-ANI measured EAS Size Spectra in comparison with KASCADE data. Both spectra for Sec uniform zenith angles interval are shown.. The line indicate EAS size attenuation versus of slant depth of atmosphere. The attenuation length by MAKET-ANI data is equal Λ=211±38 g/cm 2, for KASCADE data Λ=197±13 g/cm 2 and by joint analyses - Λ=194±14 g/cm 2. The difference of spectral indexes before and after knee is equal Δγ = - 0.45±0.06 EAS size spectra approximated by

8. L D F The comparison the experimentally obtained LDF functions to the theoretical ones: CORSIKA 562 (QGSJet01, NKG mode) was used, for H, He, O, Si, Fe primaries 10 6 events and E 0 ≥10 5 Gev for each of them generated we assume the following mass composition and energy spectra of the primary galactic cosmic rays (so called ”normal” mass composition): (35%H, 25%He, 14%O, 15% Si, 10% Fe); knee position E knee = Z ∙ E 0, E 0 = 3 ∙ 10 15 eV, Z is the primary nuclei charge; energy spectra index γ 1 = -2.7 before knee and, γ 2 = -3.1 after knee for all nuclei.

9. The dependence of the measured age parameter on shower size in comparison to several models of primary flux composition is presented To outline the boundaries of the extreme cases we use pure proton and pure iron nuclei fluxes. More realistic assumptions about energy dependence of the primary composition are between these extreme assumptions: 1. Rigidity dependent “normal” composition with knee position at E knee = Z ∙ 3 ∙ 10 15 eV 2. The same as in point 1, but with fixed knee position at E knee = 3 ∙ 10 15 eV for each group of nuclei 3. "Heavy " composition (5% P, 5% He, 10% O, 10% Si, 70% Fe), knee position at E knee = Z ∙ 3∙ 10 15 eV. For all 3 models the energy spectra index γ 1 = -2.7 before knee and, γ 2 = -3.1 after knee for all nuclei. The first model (“normal” composition) fits experimental data quite well, (the value of the test equals χ 2 =1.23). At the same time we can exclude the options 2 (χ 2 =8.8) and 3 ( χ 2 = 31.1 ).

10. Primary CR simulation by CORSIKA (562,QGSJet01,NKG mode) code was done. For H, He, O, Si, Fe primaries 10 6 events and E 0 ≥10 5 GeV for each of them generated. The atmosphere depth is 700g/cm 2 Yield function from simulated primaries after:  trigger conditions;  response function of installation, is obtained. The response of the MAKET- ANI detector on simulated “light” (p+He ) and “heavy” (Si+Fe) induced showers, versus primary energy; shower selection criteria: N e ≥10 5, θ≤30 o (closed symbols ), N e ≥10 5, 30 o ≤θ≤45 o (open symbols)

12. Primary light component (p+He) measured by the MAKET-ANI detector with comparison to the results from KASCADE, EAS-TOP, HEGRA, EAS- TOP+MACRO, TIBET and primary protons spectra approximations obtained by the single hadrons fluxes EAS-TOP and KASCADE. (All data based on CORSIKA QGSJet01 version). The direct balloon measurements by ATIC-2 and JACEE at 10 2 – 10 5 GeV also presented.

13. The energy spectrum of the “heavy” nuclei group measured by the MAKET-ANI detector along with spectra from KASCADE, EAS- TOP+MACRO and ATIC-2. The solid line is a power function approximations

14. SPECTRUM OF COSMIC RAYS, PRODUCED IN SUPERNOVA REMNANTS E.G. Berezhko, H.J.Völk April 16, 2007

15. Conclusions. From 1997 up to the end of 2004 the MAKET – ANI experiment has taken data with exposition time of ~1.42∙10 8 s. The total number of the registered shower events was ~1.2∙10 7. A smaller sample of the data (~1.3∙10 6 ) with Ne ≥10 5 and θ ≤46.8 0 was used for the in-depth analysis of the LDF and size spectra. By 7.2∙10 5 near the vertical (θ ≤30 0 ) EAS were obtained the energy spectra of light and heavy nuclei groups. The obtained dependence of the shower age on shower size pointed to the weighting of the primary flux mass composition after the knee of the “all particle” spectrum; The size spectra show evidence of a “knee” at shower size ~10 6 particles. As the zenith angle enlarges, the knee position is moving to smaller sizes, according to the EAS attenuation length =211±38 g/cm 2.

16. The difference of the power low spectra before and after the knee is constant with high precision Δγ = 0.45 ±0.06. The estimated energy spectrum of the light mass group of nuclei shows a very sharp knee: ∆γ ≈ 0.9, compared to ∆γ ≈ 0.4 for the all-particle energy spectra. The energy spectrum of the heavy mass group of cosmic rays shows no knee in the energy interval of 10 15 –10 16 eV. The mentioned results are consistent with the nonlinear kinetic theory of CR acceleration in SNR shells