Charged-particle and gamma-ray astronomy: deciphering charged messages from the world s most powerful

Transcription

1 Charged-particle and gamma-ray astronomy: deciphering charged messages from the world s most powerful Charged-particle astronomy coming of age How it is done The sources The signals What we have learned 1

2 Cosmic accelerators 2

3 Fermi acceleration in AGN and radio galaxies 3

4 Fermi acceleration + B E 1 E 2 4

5 Fermi acceleration: simple estimates Let us assume that each crossing give the particle acceleration by factor β, and the probability to remain in the acceleration region is P. Then the number of particles that cross the shock front k times is N = N 0 P k, and their energy is E = E 0 β k Eliminate k to get N/N 0 = (E/E 0 ) lnp/lnβ, or dn de = E α, where α = 1 lnp lnβ Hence, the power-law spectrum. 5

6 Fermi acceleration: simple estimates Let us assume that each crossing give the particle acceleration by factor β, and the probability to remain in the acceleration region is P. Then the number of particles that cross the shock front k times is N = N 0 P k, and their energy is E = E 0 β k Elimitae k to get N/N 0 = (E/E 0 ) lnp/lnβ, or dn de = E α, where α = 1 lnp lnβ Hence, the power-law spectrum. 6

7 Detection: Pierre Auger Observatory 7

8 Little known fact about lightnings 8

9 Pierre Auger in Malargue, Argentina 9

10 Pierre Auger Observatory 10

11 11

12 Water tanks 12

13 Water tanks 13

14 14

15 Fluorescence detector 15

16 Fluorescence detector 16

17 Lidar is used to monitor the atmosphere 17

18 Events PRELIMINARY 18

19 19

20 Extreme Universal Space Observatory (EUSO) EUSO is designed to observe fluorescent air showers initiated by extremely high energy cosmic rays - and neutrinos 20

21 EUSO is designed to observe fluorescent air showers initiated by extremely high energy cosmic rays - and neutrinos 21

22 A recent discovery by Pierre Auger energy-dependent chemical composition ] 2 <X max > [g/cm QGSJET01 QGSJETII Sibyll2.1 EPOSv1.99 proton ] 2 ) [g/cm max RMS(X proton iron iron E [ev] E [ev] Why should any source accelerate nuclei rather than protons at high energies?? 22

23 Composition puzzle It is hard to imagine an extragalactic source that accelerates selectively heavy nuclei. In fact, the opposite is expected from photodissociation of nuclei at the source. 23

24 Composition puzzle It is hard to imagine an extragalactic source that accelerates selectively heavy nuclei. In fact, the opposite is expected from photodissociation of nuclei at the source. If sources accelerate all the particles, can propagation effects alter the observed composition? 23

25 Composition puzzle It is hard to imagine an extragalactic source that accelerates selectively heavy nuclei. In fact, the opposite is expected from photodissociation of nuclei at the source. If sources accelerate all the particles, can propagation effects alter the observed composition? Yes, if the sources are Galactic. Diffusion times for nuclei are longer than for protons of the same energy. ( ) t D R2 R 2 ( 26 D 108 yr 10 kpc Z ev E ) 2 Diffusion is also energy-dependent. 23

26 Galactic sources likely to exist... in the past Long list of candidates (in possibly overlapping categories): GRBs hypernovae collapsars other unusual supernovae 24

27 Galactic sources likely to exist... in the past Long list of candidates (in possibly overlapping categories): GRBs hypernovae collapsars other unusual supernovae Do these events occurred in our our own galaxy? 24

28 Hypernova observed in our Galaxy 25

29 GRBs as sources in Milky Way Galaxy GRBs have been proposed as sources of extragalactic UHECRs [Vietri; Waxman; Dermer (cf. talks by Dermer, Waxman)] Galactic GRBs have been considered as sources of UHECRs [Dermer et al., Biermann et al.] Long GRBs: probably unusual supernova explosions. Short GRBs: probably mergers of compact stars. Both should have happened in our own Galaxy in the past, at a rate of one per years. Past Galactic GRBs have been considered as the explanation of 511 kev line from the Galactic Center [Bertone, et al.; Parizot et al., Calvez, AK], as well as the electron excess of PAMELA/Fermi [Ioka; Calvez, AK] How long will the UHECRs diffuse in the Galactic magnetic fields, and how isotropic will they become? Depends on composition. 26

30 Transport equation (a simple exercise) n i t (D i n i ) + E (b in i ) = Q i (E, r,t) + P ik (E,E )n k (E )de. k For energies below GZK cutoff, neglect energy losses; consider just diffusion. For a pointlike source ( ) γ E0 Q i (E, r,t) = δ( r)q 0 E the solution is n i (E,r) = Q 0 4πr D i (E) ( E0 E ) γ. 27

31 Diffusion in two different regimes R <<l ; deflect. M.F.P.=l B B c l B l c R >>l ; deflect. M.F.P.>>l B c D(E)=D =l/3, for E<<E D(E)=D (E/E ), for E>>E critical energy, E : R =l 0 B c 28

32 Diffusion in two different regimes R <<l ; deflect. M.F.P.=l B B c l B l c R >>l ; deflect. M.F.P.>>l B c D(E)=D =l/3, for E<<E D(E)=D (E/E ), for E>>E critical energy, E 0: R =l B c E 0 depends on Z of a nucleus. 28

33 Diffusion in two different regimes R <<l ; deflect. M.F.P.=l B B c D(E)=D =l/3, for E<<E l B l c R >>l ; deflect. M.F.P.>>l B D(E)=D (E/E ), for E>>E critical energy, E : R =l E 0,i depends on Z i of a nucleus i. c 0 B c Critical energy E 0,i, at which R B,i = l c, depends on Z i R i = E ( ) E = l 0, where Bq i E 0,i E 0,i = ebl 0 Z i, E 0,i = Z i ev ( B G )( l0 0.3 kpc ) 29

34 Diffusion in two different regimes: D i (E) = What about our solution? ( ) δ1 E D 0 E, E E0,i, ( 0,i ) (2 δ2 ) E D 0 E, E > E0,i. 0,i n i (E,r) = 4πr D i (E) E The spectral slope changes at E E 0,i, and the flux drops dramatically because the particles escape from the galaxy. The flux drops for protons at lower energies than heavy nuclei. Q 0 ( ) γ E0 30

35 Energy-dependent composition due to diffusion protons, C, Fe E 3 dn de GeV 2 km 2 s 1 sr E GeV 31

36 More realistic source distribution Supernovae or long GRBs, assuming they follow star counts [Bahcall et al.] Short GRBs, based on observed distribution in other galaxies [Cui, Aoi, Nagataki] 32

37 Comparison with Pierre Auger data æ æ E dn/de (ev km s sr ) p æ æ æ æ æ Fe æ æ æ æ æ æ æ æ E (ev) 10 [Calvez, AK, Nagataki, PRL 10] Energy in UHECR per source (GRB, hypernova, etc.) is erg above ev. 33

38 Galactocentric anisotropy (sources follow stars) p δ Fe Total E (ev) [Calvez, AK, Nagataki]

39 Clusters of events from recent/closest GRBs supernovae/long GRBs short GRBs In addition, extragalactic protons can show correlation with distant sources. 35

40 Which story is true? Galactic sources energy-dependent composition with nuclei dominating at ev Galactocentric anisotropy 2% extragalactic sources contribute in protons > ev. any correlation of extragalactic protons with sources is diluted by the uniform Galactic nuclei Galactic nuclei cluster in directions of closest sources No Galactic sources protons dominate at high energy small-scale anisotropy in UHECR arrival directions determined by sources (and intervening magnetic fields) no Galactocentric anisotropy 36

41 Cosmic acelerators going off every so often in our own Galaxy? We are looking into the possible connection of Galactic GRBs with mass extinctions. 37

42 Acceleration in AGN and radio galaxies when AGN jet points at Earth, called blazar 38

43 39

44 Gamma-ray telescopes 40

45 Observations of distant blazars: 1ES (z = 0.14) and 3C66A (z = 0.44) ] -1 TeV -1 s -2 dn/de [ cm dn/de (GeV cm s ) Energy [ TeV ] E(GeV) HESS (black), MAGIC (blue) and VERITAS (red) data points 41

46 Observations of distant blazars: 1ES (z = 0.14) and 3C66A (z = 0.44) ] -1 TeV -1 s -2 dn/de [ cm dn/de (GeV cm s ) Energy [ TeV ] E(GeV) HESS (black), MAGIC (blue) and VERITAS (red) data points Surprise: no signs of absorption due to γγ EBL e + e! 41

47 Possible solution: 42

48 A one parameter fit, 3C66A (param = the power of ANG in CR, subject to constraints) dn/de (GeV cm s ) [Essey, AK] E(GeV) 43

49 Something else we ve learned: IGMF Gauss Intergalactic magnetic fields make the images of blazars diffuse. We have discovered halos around blazars, which imply Gauss mgnetic fields [Ando, AK]. Rel. DEC (deg) (a) Counts Map (3-10 GeV) Rel. DEC (deg) (b) Model Map (3-10 GeV) (c) Counts Map ( GeV) Rel. DEC (deg) (d) Model Map ( GeV) Rel. DEC (deg) Rel. RA (deg) Rel. RA (deg) Rel. RA (deg) Rel. RA (deg) (counts) (counts) (counts) (counts) Small fields, small correlation length We are probably seeing the seed fields from the Big Bang! [Ando, AK, ApJL] (counts) (counts) 44

50 Alexander Kusenko (UCLA) Toronto 10 Simple analogy To see the halo at night, one could stack the images of stars. 45

51 Something else we ve learned: IGMF Gauss Intergalactic magnetic fields make the images of blazars diffuse. We have discovered halos around blazars, which imply Gauss mgnetic fields [Ando, AK]. Rel. DEC (deg) (a) Counts Map (3-10 GeV) Rel. DEC (deg) (b) Model Map (3-10 GeV) (c) Counts Map ( GeV) Rel. DEC (deg) (d) Model Map ( GeV) Rel. DEC (deg) Rel. RA (deg) Rel. RA (deg) Rel. RA (deg) Rel. RA (deg) (counts) (counts) (counts) (counts) Small fields, small correlation length We are probably seeing the seed fields from the Big Bang! [Ando, AK, ApJL] (counts) (counts) 46

52 Conclusion Deciphering the messages from the most powerful objects in the universe has taught us about the acceleration of cosmic rays and gamma rays, intergalactic magnetic fields gigantic explosions in our own Galaxy Undoubtedly, more discoveries ahead! 47