1. Interpretation of the Cosmic-ray Energy Spectrum and the Knee Inferred from the Tibet Air-Shower Experiment M.Shibata* and Tibet ASg Collaboration *Yokohama National University ICRC2009 Lodz, Poland

2. The Tibet AS  Collaboration M.Amenomori(1), X.J.Bi(2), D.Chen(3), S.W.Cui(4), Danzengluobu(5), L.K.Ding(2), X.H.Ding(5), C.Fan(6), C.F.Feng(6), Zhaoyang Feng(2), Z.Y.Feng(7), X.Y.Gao(8), Q.X.Geng(8), Q.B.Gou(2), H.W.Guo(5), H.H.He(2), M.He(6), K.Hibino(9), N.Hotta(10), Haibing Hu(5), H.B.Hu(2), J.Huang(2), Q.Huang(7), H.Y.Jia(7), L.Jiang(8, 2), F.Kajino(11), K.Kasahara(12), Y.Katayose(13), C.Kato(14), K.Kawata(3), Labaciren(5), G.M.Le(15), A.F.Li(6), H.C.Li(4, 2), J.Y.Li(6), C.Liu(2), Y.-Q.Lou(16), H.Lu(2), X.R.Meng(5), K.Mizutani(12, 17), J.Mu(8), K.Munakata(14), A.Nagai(18), H.Nanjo(1), M.Nishizawa(19), M.Ohnishi(3), I.Ohta(20), S.Ozawa(12), T.Saito(21), T.Y.Saito(22), M.Sakata(11), T.K.Sako(3), M.Shibata(13), A.Shiomi(23), T.Shirai(9), H.Sugimoto(24), M.Takita(3), Y.H.Tan(2), N.Tateyama(9), S.Torii(12), H.Tsuchiya(25), S.Udo(9), B.Wang(2), H.Wang(2), Y.Wang(2), Y.G.Wang(6), H.R.Wu(2),L.Xue(6), Y.Yamamoto(11), C.T.Yan(26), X.C.Yang(8), S.Yasue(27), Z.H.Ye(28), G.C.Yu(7), A.F.Yuan(5), T.Yuda(9), H.M.Zhang(2), J.L.Zhang(2), N.J.Zhang(6), X.Y.Zhang(6), Y.Zhang(2), Yi Zhang(2), Ying Zhang(7, 2), Zhaxisangzhu(5) and X.X.Zhou(7) (1)Department of Physics, Hirosaki University, Japan. (2)Key Laboratory of Particle Astrophysics, Institute of High Energy Physics, Chinese Academy of Sciences, China. (3)Institute for Cosmic Ray Research, University of Tokyo, Japan. (4)Department of Physics, Hebei Normal University, China. (5)Department of Mathematics and Physics, Tibet University, China. (6)Department of Physics, Shandong University, China. (7)Institute of Modern Physics, SouthWest Jiaotong University, China. (8)Department of Physics, Yunnan University, China. (9)Faculty of Engineering, Kanagawa University, Japan. (10)Faculty of Education, Utsunomiya University, Japan. (11)Department of Physics, Konan University, Japan. (12)Research Institute for Science and Engineering, Waseda University, Japan. (13)Faculty of Engineering, Yokohama National University, Japan. (14)Department of Physics, Shinshu University, Japan. (15)National Center for Space Weather, China Meteorological Administration, China. (16)Physics Department and Tsinghua Center for Astrophysics, Tsinghua University, China. (17)Saitama University, Japan. (18)Advanced Media Network Center, Utsunomiya University, Japan. (19)National Institute of Informatics, Japan. (20)Sakushin Gakuin University, Japan. (21)Tokyo Metropolitan College of Industrial Technology, Japan. (22)Max-Planck-Institut fur Physik, Deutschland. (23)College of Industrial Technology, Nihon University, Japan. (24)Shonan Institute of Technology, Japan. (25)RIKEN, Japan. (26)Institute of Disaster Prevention Science and Technology, China. (27)School of General Education, Shinshu University, Japan. (28)Center of Space Science and Application Research, Chinese Academy of Sciences, China.

3. 1.Tibet air shower array. All-particle spectrum ApJ 678 (2008) 1165 2.Hybrid experiment using AS core detector to measure proton and helium spectra. Phys. Lett. B 632 58-64 (2006) & ICRC2007, 2 (2007) 121 3. Compilation of composition measurements. 4.Extra component at the knee and its origin. 5.Next phase experiment. Contents

4. Tibet IIIAS array: 700 ch x 0.5 m 2 scint. detectors with 7.5m sp. Area 37,000 m 2 Location Tibet, Yangbajing, China 4300 m a.s.l. 606 g/cm 2

5. All particle spectrum. Knee at 4 PeV dJ/dE ∝ E -γ γ=2.65  3.1 ApJ 678 (2008) 1165

6. Burst detectors

7. Emulsion Chamber and Burst Detector 2cm Artificial Neural Network N γ, ΣE γ, ＜ R γ ＞, ＜ ER γ ＞, N e, θ

8. P, He by Tibet Experiment Phys. Lett. B 632 58-64 (2006) with 30% model dependence 3.01±0.11 3.05±0.12

9. N e >3x10 5 Zenith angle θ<25 deg. Number of selected events :1176 during live time of 434.3 days. Any 3 PD signals > 100 particles equivalent after the attenuation inside scint. (N b ～ 2 x 10 4 at the center of scinti.) N b top >5x10 4, contained events 632 P,He-like events Second phase to measure P+He with higher statistics ICRC2007, 2 (2007) 121

10. Proton+Helium spectrum Phase I Phase II ICRC2007, 2 (2007) 121

11. Proton+Helium spectrum Phase I Phase II

12. PHe C O Ne Mg Si S Ar Ca SubFe Fe Fit for direct observations 10 9 10 15

13. Fit for direct observations (<100TeV)

14. Proton Spectrum Direct measurement and Tibet combined

15. ε b : break point (7x10 14 eV for proton) Δγ: difference of power index before and after the break point （ Δγ ＝ 0.4 ） Broken power law formula to describe proton spectrum

16. Multiple source model Distribution of acceleration power of cosmic rays See Poster 295 (M.Shibata) ε m ≡ ε b

17. Interpretation of ε b Minimum acceleration limit for CR protons. Threshold of SN explosion by massive stars. (type II SN by >8M ○ ）

18. CR composition at the knee ε z = Z x ε b, Δγ ＝ 0.4 700TeV

19. CASA/MIA HEGRA KASCADE DICE BASJE TIBET All particle spectrum around the knee 10 14 10 16 10 18

20. All data agree if we apply energy scale correction within 20% by normalizing to direct observations. Extra component can be approximated by suggesting nearby source(s). Since P and He component do not show the excess at the knee, the extra component should be attributed to heavy element such as Fe. (W.Bednarek and R.J.Protheroe,2002,APh) Extra component

21. P +He spectrum does not show excess at the knee Expected by multiple source model

22. Chemical composition of SN ejecta (Nomoto,K et al. Nucl. Phys. A, 621, 467, 1997)

23. If nearby type Ia SN ejecta makes knee sharp ….. All He Fe O C+N Ca Si P

24. Next phase of Tibet hybrid exp. YAC:Yangbajing Air shower Core detector MD:Muon Detector Measure the energy spectrum of the main component at the knee. Detector ： Low threshold BD grid ＋ AS array + Muon detector. Observe energy flow of AS core within several x 10m from the axis.

25. 2m underground, 20 units, 9000 m 2 Water Cherenkov Muon Detector (MD)

26. Summary Proton and helium spectra at the knee measured by Tibet hybrid experiment show steep power index of around 3.1 and low fraction to the all particles. Systematic error due to the interaction model dependence is within 30% for the flux. Broken power law spectrum is used to summarize the chemical composition measurements based on multiple source model. All particle spectrum in wide energy range around the knee shows presence of an extra component which mainly consists of heavy elements and its spectrum suggests the contribution of nearby source(s). Next phase of Tibet experiment, Tibet III+YAC+MD, will measure the heavy component at the knee and also measure γ-ray spectrum with p/γ separation of AS.

27. Thank you

28. Pb 7cu Iron Scint. Box Design of YAC 40cm x 50cm, 20x20 channels S=5000m 2 3.75m spacing 400ch N b >100, any 5 (>30GeV) Wave length shifting fiber + 2 PMTs (Low gain & High gain) 10 2 <N b <10 6

29. All-particle Spectrum in Wide Energy Range (10 14 -10 17 eV) Energy determination : Lateral Distribution Fitting using modified NKG function Derivation of the function was made by detailed detector Monte Carlo (Corsika & EPICS) Carpet array calculation (lateral structure, total size) Sampling array calculation (fit size, resolution) Longitudinal age parameter output by Corsika is used to describe the structure function.

30. a(s) b(s) Modified NKG function r m ’=30 m S-2 S-4.5

31. For pure electromagnetic cascade r m ’ = 80m a(s) b(s)

32. Size resolution by reconstructing MC events

33. Energy resolution ~36.0% (about 150-250TeV) Energy resolution ~16.9% (about 1500-2500 TeV) Energy resolution ~11.1% (about 6×10 4 - 8×10 4 TeV) Primary energy resolution

34. Size spectrum of Tibet III

35. Generation efficiency of  family event by primary protons in QGSJET and SIBYLL QGSJET SIBYLL SIBYLL/QGSJET ~1.3 SIBYLL/QGSJET ～ 1.3 SIBYLL QGSJET SIBYLL QGSJET 10 14 10 15 10 16 E 0 eV 10 14 10 15 10 16 E 0 eV

36. Artificial Neural Network JETNET 3.5 Parameters for training: N γ, ΣE γ, ＜ R γ ＞, ＜ ER γ ＞, N e, θ

37. Primary proton spectrum Preliminary (KASCADE data: astro-ph/0312295) All Proton KASCADE (P) Present Results (a) ( by QGSJET model) (b) ( by SIBYLL model )

38. Primary helium spectrum (a) (by QGSJET model)(b) (by Sibyll model) p+helium selection: purity=93%, efficiency=70%

39. J.R.Hoerandel, Astroparticle Phys. 21,241-265(2004) QGSJET SIBYLL KASCADE

40. Fraction of elements (%) 141516

41. Helium dominant composition by Kascade e-μmeasurement Kascade QGSJET spectrum E b =4 PeV for P

42. Simulation (Phase I) Corsika 6.030 QGSJET01,SIBYLL2.1 (high energy int. model) ｘ Heavy Dominant Composition (HD) Proton Dominant Composition (PD) ＝ analyses under 4 models Event Selection AS size Ne>2 x 10 5 accompanied by γ family of E γ >4TeV, n γ 4, ΣE γ >20 TeV

43. Tibet Hybrid Experiment Tibet As γ Collaboration 1996 ー 1999 AS+EC+BD ～ 200 events  P,He spectrum Phase2:2002 ー 2005 AS+BD Light component(P+He) with high statistics>1000 events Phase3:in preparation AS+BD grid array Observe heavy component at the knee

44. Contribution of nuclei with odd Z is corrected using solar abundance

45. The sharpness of the knee suggests limited range of atomic numbers for extra component

46. Comparison with chemical composition of Type Ia SNR ejecta

47. Ca Expected spectrum of heavy components Si S Fe

48. Average Mass GCR+EXGCR (mixed comp.) GCR only EXGCR=P

49. Can CNO constitute extra component?

50. Akeno No correction Normalized at low energy Normalized at high energy

51. Anisotropy Σρ>1000 （ E 0 ＞ 10 14 eV ） Orion complex l=205.5 b=+0.5 Monoceros_Nebula （ α= 99.750 δ=6.500 ） SNR Monoceros is colliding with Rossete nebula. EGRET (98.276 6.764 GEVJ0633+0645)

52. Initial mass function (IMF) ∝ M -2.5

53. Stellar life time ∝ M -1 ∝ M -3.8

54. ? Relation between Fermi e ± and extra component at the knee? PeV nuclei + target （ ＊ 10TeV/n) π 0 γ ＊ 100GeV e ± This may be quite possible scinario, but not calculated yet.