Extragalactic Background Light and Gamma-ray Blazars

Transcription

1 Extragalactic Background Light and Gamma-ray Blazars Yoshiyuki Inoue (JAXA International Top Young ISAS/JAXA) with thanks to A. Dominguez, S. Inoue, K. Ioka, O. E. Kalashev, J. Kataoka, A. Kusenko, M. Kobayashi, G. Madejski, R. Makiya, M. Mori, Y. Niino, R. Sato, T. Totani, Y. Takahashi, Y. Tanaka, E. Wright 1

2 Cosmic Background Radiation E 2 dn/de [erg cm -2 s -1 sr -1 ] CMB Galaxies (Inoue et al. 13) Pop-III Stars (Inoue et al. 13) AGNs (All) Radio-quiet AGNs (Inoue et al. 08) Blazars (Inoue and Totani 09) Radio Galaxies (Inoue 11) Galaxies? AGNs? Photon Energy [ev] 2

3 Extragalactic Background Light (EBL) I [nw/m 2 /sr] 0 1 Stars Dust Madau & Pozzetti 00 (HST) Elbaz et al. 02 (ISO) Papovich et al. 04 (Spitzer) Fazio et al. 04 (Spitzer) Xu et al. 05 (GALEX) Dole et al. 06 (Spitzer) Frayer et al. 06 (Spitzer) Gardner et al. 00 (HST) Berta et al. 11 (Hershel/PEP) Wright & Reese 00 (DIRBE) Wright 04 (DIRBE) Levenson et al. 07 (DIRBE) Levenson & Wright 08 (DIRBE) Bernstein 07 (HST) Matsuoka et al. 11 (Pioneer) Matsumoto et al. 11 (IRTS) Matsuura et al. 11 (AKARI) CambrØsy et al. 01 (DIRBE) Dwek & Arendt 98 (DIRBE) Gorijian et al 00 (DIRBE) Finkbeiner et al 00 (DIRBE) Hauser et al 98 (DIRBE) Lagache et al 00 (DIRBE) Edelstein et al 00 (Voyger) Brown et al 00 (HST/STIS) Albert et al 08 (MAGIC) (X, z c )=(1.0, 0.0) (X, z c )=(50.0,.0) (X, z c )=(0.0,.0) [µm] 3 YI+ 13a

4 Gamma rays are attenuated by EBL blazar Extragalactic Background Light IACT γ VHE γ EBL e + e - 4

5 EBL Constraints from Gamma Rays LAT best fit -- 1 sigma LAT best fit -- 2 sigma Franceschini et al Finke et al model C Stecker et al High Opacity Stecker et al Low Opacity Kneiske et al highuv Kneiske et al best fit Kneiske & Dole 20 Dominguez et al Gilmore et al fiducial Abdo et al. 20 ] sr ME12 exclusion region H.E.S.S. H.E.S.S. low energy H.E.S.S. full dataset H.E.S.S. high energy H.E.S.S. contour (sys + stat ) low full high Direct measurements τ γ γ 1 Fermi [ nw m F λ λ -1 z~1.0 z Galaxy counts 2 Energy [GeV] Ackermann + 13 Fermi and H.E.S.S. derived the EBL opacity or intensity using the combined spectra of blazars (see also Gong & Cooray 13, Dominguez + 13). Inconsistent with the NIR EBL excess. They assume simple log-parabola or power-law spectra. 5 1 λ [ µm ] Abramowski + 13 ig. 5. Flux density of the extragalactic background light versus wave-

6 Direct Measurement of EBL Pioneer /11 IRTS AKARI Matsuoka + 11 Tsumura + 13 Pioneer /11 measurements are consistent with galaxy counts. Recent AKARI measurements are consistent with IRTS. EBL peak at near infrared? 6

7 >0 GeV Gamma Rays from z=1.1 & z>0.6 PKS at z=1.1 PKS at z > Observed EBL corrected (Franceschini et al. 2008) EBL corrected (Inoue et al. 2013) Broken PL model Contemporaneous Fermi LAT from Acciari et al. 20 Contemporaneous Fermi LAT Power-law Fit from Acciari et al. 20 VERITAS Observed VHE Spectrum from Acciari et al. 20 E 2 dn/de [erg cm -2 s -1 ] - (Jy Hz) νf ν z= Absorption-corrected Spectrum using Doming uez et al z= Absorption-corrected Spectrum using Gilmore et al z=1.2 Absorption-corrected Spectrum using Dominguez et al Fermi 11 Fermi + VERITAS Energy [GeV] Tanaka, YI, Energy (GeV) Furniss Distant very high energy (VHE) sources show spectral hardening and do not show short time variabilities. 7

8 Is VHE Spectral Hardening Universal? E 2 dn/de [TeV/cm 2 /s] -9 - H z = ES z = S z = 0.31 RX J z = RBS 0413 z = 0.19 Spectra of blazars at z > 0.15 show hardening from a few hundred GeV. 1ES z = ES z = ES z = ES z = Energy [TeV] 4C z = Energy [TeV] 3C 66A z = Energy [TeV] 3C 279 z = Energy [TeV] YI+ 13a 8 Γ z z Essey Γ & Kusenko = Γ 12Γ

9 Secondary Gamma Rays? Stochastic Acceleration? KUV (z=0.61) Secondary Gamma Rays E 2 F E [erg cm -2 s -1 ] γ-induced (low IR) γ-induced (best fit) CR-induced (low IR) CR-induced (best fit) Becherini et al. (2012) H.E.S.S. I CTA 1ES (z=0.1396) Stochastic Acc. log vf 2 v [erg.sec.cm2 ] F v v 1/3 1ES (z=0.1396) E 2 Exp[-(E/E c ) 3 ] H.E.S.S. low level EBL - E 2 Exp[-E/E c ] H.E.S.S. high level EBL SWIFT -.5 F v v 1/ Secondary gamma rays from cosmic rays along line of sight (Essey & Kusenko, Essey+, Essey+ 11, Murase+ 12, Takami+ 13). CTA will statistically test this scenario through logn-logs (YI + 14) E [ev] Stochastic acceleration can generate hard electron spectra (Stawarz & Petrosian 08, Lefa+ 11). Takami log v [Hz] = Lefa+ 11

10 No. ] Can we explain the NIR EBL excess in spectrum and fluctuation? Extragalactic Background Light Spectrum with AKARI IRC 5 Figure 2. Amplitude maps of the two-dimensional fluctuation spectra, [q 2 P2 (q)/(2π )]1/2, in Fourier space. The results for the 2.4, 3.2, and 4.1 µm band are shown from the left to the right. The grayscale bars below each map indicate amplitudes of the fluctuation in units of nw m 2 sr 1. AKARI IRTS AKARI Matsumoto + 11 Figure 3. Upper panel shows the one-dimensional fluctuation spectra, [q 2 P2 /(2π )]1/2 in units of nw m 2 sr 1, obtained by two-dimensional Fourier analysis as a function of angular scale (2π )/q. Graphs correspond to the 2.4, 3.2, and 4.1 µm bands from left to right. Filled circles and open triangles show the fluctuation spectra for sky and dark maps, respectively. The lower panel shows the fluctuation spectra of the sky after subtracting those of dark maps in quadrature. The straight lines fluctuation spectra of shot noise due to unresolvedfrom faint galaxies. Allincluding error bars represent 1σ error. showindicate EBLtheby various direct photometry space Tsumura + 13 Fig. 4. Spectrum of EBL and integrated light of galaxies. Filled plots this study, and open plots shows the integrated light of galaxies by deep observations. Horizontal bars show the band widths of results taken as sequence alternating time, observed while in case B the wide-band data. Solid curve shows a model spectrum of the integrated light of agalaxies based onin the evolution of thefor both subsets are consistent with those of the stacked dark maps. This indicates that the observed structure is indeed to the earlier curve and latershows halves a of scaled our set. version rest-frame K-band galaxy luminosity function up to redshift 4 (Domı nguez subsets et al. correspond 2011), and broken of present in the original images and is of celestial origin. For both cases, we obtained two stacked images, F1 and F2, by it in case of AKARI s detection limit of point sources (mk = 19). We also examined the impact of masking on the fluctuation applying the same procedure as the one described previously in spectra, since 53% of all pixels are masked. We constructed this section. The fluctuation spectra of the difference between a common mask that includes all pixels masked in any of three twobe stacked images arealready, shown in Figure Cases A and must detected if it4. exists. An isotropic his correlation to the higher Galactic latitude regions in Earth these B are shown as squares and asterisks, respectively, while the anotherbands. The fraction of remaining pixels in images ur method. However, this assumption is obviously too diffuse background from the Oort cloud could be wavelength with the common mask applied is 32%. We again performed fluctuation spectra of the dark maps are shown by triangles. The imple. For example, UV radiation field at high Galactic atitude is weaker than that at Galactic plane (Seon et l. 2011), therefore the PAH molecules are less excited at candidate. However, the very blue spectrum toward 1 µm cannot be generated by thermal emission from very 4 cold dust (<30 K) at the Oort cloud. Scattered sunlight

11 First stars? Lyman alpha photons from z~ will redshifted to ~1 um at z=0. We might see the light for first stars. But, we need very high first star formation rate density. Fernandez & Komatsu Fernandez & Komatsu 06

12 Reionization Constraints YI+ 13 Neutral Hydrogen Fraction YI+ 13 Electron Thomson scattering opacity Ionizing photon emissivity of first stars can not violate these observed reionization data. 12

13 Constraints on First Stars Cosmic Star Formation Rate Density [M sun /yr/mpc 3 ] Fernandez & Komatsu 06 Raue, Kneiske, & Mazin 09 Gilmore 12 C = 3.0, f esc = 0.2 C = 3.0, f esc = 0.5 C = 1.0, f esc = 0.2 Bromm & Loeb 06 Trenti & Stiavelli 09 de Souza, Yoshida, & Ioka Redshift z YI+ 14 Combining reionization and distant gamma-ray data (E<0 GeV). The required first star formation rate density is inconsistent with reionization data (e.g. Madau & Silk 05; YI+ 14) 13

14 Semi-analytical Galaxy Formation Model with First Stars YI+ 13a YI+ 13a A galaxy formation model including first stars which is consistent with reionization data. Pop-III contribution is <0.5% of total NIR EBL. 14

15 LETTER RESEARCH Intracluster Halo Stars? a l 2 C l /2π (nw m 2 sr 1 ) 2 b π/l (arcmin) IHL 1-halo 2-halo Shot noise l 3.6 μm Low-z z > 6 2π/l (arcmin) Stars stripped from host galaxies by major mergers. Intrahalo stars may create a fluctuation peak at l~00. Is this population already taken into account in galaxy counts? 0 1 Ref. 14 l 2 C l /2π (nw m 2 sr 1 ) μm f IHL 1 2 Ref. 13 Ref. 25 This study M31 MW l Cooray μm Halo mass (M/M ) Cooray+ 12

16 Can we explain the AKARI measurement? IRTS 1) Origins should be born not before the reionizaiton. AKARI 2) Origins should have effective temperature around 4 K. 3) Origins should not be galaxies. Tsumura

17 Intergalactic Medium (IGM) and EBL IGM are heated up to 4 K. IGM are kept ionised ~98.3% of baryons are ionised (Fukugita & Peebles 04). most of baryons are not in galaxies but in IGM. Springel+ 05 IGM are inhomogeneously distributed. 17

18 IGM Density Distribution Current baryon mean density : Overdensity: IGM density evolution follows the structure formation. Miralda-Escude, Haehnelt, & Rees 00 We adopt Miralda-Escude, Haehnelt, & Rees 00 up till z=0. 18

19 Thermal States of IGM! cooling radiative heating Valageas et al. 02 Fig. 1. The phase-diagram of the IGM from z = 3 down to z = 0. The straight solid line Tα cool IGM (Lyman-α forest). curved line Tgh shows, a mean trend,etthe (e.g. The Bolton et al.solid 08, Furlanetto & Ohas 08, McQuinn al. Equa which is shock-heated through the building of non-linear gravitational structures. The das around the allowed domain for this warm IGM. In counter-clockwise order, starting from 1) a fast-cooling region where the gas cannot remain over a Hubble time, 2) a high-density an the exponential tail of the pdf P(δR ) which only contains very rare massive halos, 3) larg shock-heating has not appeared yet, 4) a low-density region within the tail of the pdf P(δR ) voids and 5) a lower-bound for the temperature Tα set by radiative heating from the UV back the density threshold vir of just-virialized halos. The points show the results of numerical s (Fig. 11). (Lee 12) Ly alpha forest observations 09 ) suggest A few percent systematic errors in the quasar continuum will lead errors in γ of the order of unity. 19

20 Fraction of Warm-hot IGM and Cool IGM The fraction of shocked IGM (Warm-hot IGM; WHIM) increased due to the structure formation. Even at z=0, ~50% of baryons are in diffuse IGM or in halo whose temperature is ~ 4 K. Probe of missing baryons. Dave

21 IGM EBL Spectrum 0 n max = 0.1 cm 3 0 n max = 0.1 cm 3 = 0.75 I [nw/m 2 /sr] I [nw/m 2 /sr] 1 1 IGM ( = 0.70) IGM ( = 0.75) IGM ( = 0.80) Galaxies Galaxies + IGM ( = 0.75) 0.1 IGM (Total) 0 < z < < z < 1 1 < z < 2 2 < z < 3 3 < z < [µm] YI+ in prep [µm] YI+ in prep. Free-bound emission dominates. IGM at z<2 are important. Strongly dependent on the density-temperature relation. 21

22 IGM EBL Fluctuation l(l+1)c l /2 [nw 2 /m 4 /sr 2 ] IGM ( =0.75, n max =0.1 cm 3 ) IGM ( =0.80, n max =0.1 cm 3 ) IGM ( =0.70, n max =0.01 cm 3 ) IGM ( =0.75, n max =0.01 cm 3 ) galaxies AKARI Angular scale [arcsec] YI+ in prep. EBL Fluctuation at 2.4 um 0 Multipole l AKARI & Spitzer reported an excess in the NIR EBL fluctuation. Non-linear growth effect is not included. NL effect will enhance the angular power spectrum at larger l (l~3000). 22

23 Summary VHE distant sources show unexpected spectral hardening. It is not straightforward to constrain EBL through gamma-ray observations. This hardening may come from propagation of cosmicrays, stochastically accelerated electrons, or other mechanisms. First stars can not explain the NIR EBL. IGM may explain the NIR EBL. 23