Obscured AGN and XRB models Andrea Comastri INAFOABolognaItaly

Obscured AGN and XRB models Andrea Comastri (INAF-OABologna-Italy) Roberto Gilli (INAF-OABologna-Italy) F. Fiore G. Hasinger M. Brusa C. Vignali R. Della Ceca G. Malaguti (INAF-OARoma-Italy) (MPE-Garching- Germany) (MPE-Garching-Germany) (Bologna University) (INAF-OABrera-Italy) (INAF-IASFBo-Italy) Simbol. X, May 15, 2007

G Deep X-ray surveys have resolved some 80% of the XRB (below a few ke. V ) but “only” 50% > 6 ke. V and a few (1 -2%) above 10 ke. V. G Wealth of X-ray and multiwavelength followup available -> possible to predict the missing population (Compton Thick AGN) G The missing population may provide an important contribution to the BH mass function SMBH/galaxy evolution, accretion rates, Eddington ratios, … Simbol. X, May 15, 2007 2

Simbol. X, May 15, 2007 3

Frontera+07 astro-ph/0611228 Simbol. X, May 15, 2007 4

Model scheme • • • XRB flux “uncertain” , so use the more robust constraints below 10 ke. V (eg. source counts) to lock the properties (eg obs/unobs ratio, NH dist. ) of the Compton Thin (log NH < 24) AGN Estimate the Compton Thin/Unobs. Ratio by comparing hard vs. soft XLF. Absorption distribution from X-ray counts. Include continuum slope dispersion. Add Compton thick AGN to fit the 30 ke. V bump Verify assumptions/make predictions on Compton Thick AGN Simbol. X, May 15, 2007 5

Unabsorbed: log. NH<21 (plus Compton reflection) Compton-Thin: 21<log. NH<24 Compton-Thick: log. NH>24 Mildly (log NH =24 -25) (NGC 6240, Circinus) Simbol. X, May 15, 2007 Heavily (log NH >25) (NGC 1068) 6

Best fit ratios ~ 4 at low Log Lx < 44 Luminosities ~ 1 at high Log Lx > 44 luminosities Simbol. X, May 15, 2007 7

X-ray log. N-log. S Simbol. X, May 15, 2007 8

NH distribution Th in k c i h T Simbol. X, May 15, 2007 9

G Radio loud AGN / Blazars contribute some 10% to the soft XRB (Galbiati et al 2004; Giommi et al 2006) G Normalizing to 1 ke. V for a range of slopes (0. 5 -0. 7) a break or cut-off must be present around 1 Me. V otherwise the softgamma-ray background would be exceeded High energy cut-off 100 -400 ke. V Simbol. X, May 15, 2007 10

Unresolved Background AC 05 Worsley+05 hard X-ray surveys still miss the most obscured sources (do not sample the XRB peak) Missing population: (numerous) moderate luminous, log NH > 23, z = 0. 5 -2 (Worsley+04, 05, AC 04, 05) and Compton Thick AGN (log NH > 24) Simbol. X, May 15, 2007 11

Simbol. X, May 15, 2007 12

70 -75% of The Beppo. SAX/ INTEGRAL 10 -40 ke. V XRB Resolved Simbol. X, May 15, 2007

Obscured AGN fraction vs sample limiting flux All abs. Thick Simbol. X, May 15, 2007 14

Expected number of sources per SX field Gilli +07 La Franca+05 Della Ceca Simbol. X, May 15, 2007

SUZAKU OBSERVATION OF Mildly CT nearby galaxies AC+07 (astro-ph/0704. 1253) Scattering fraction < 0. 5 % NGC 5728 Scattering fraction ~ 2% See also Wilman et al. 2002 Hy. LIG NGC 4992 Simbol. X, May 15, 2007 16

4 X Simbol. X, May 15, 2007 17

Heavily Thick will not be well sampled by deep hard X-ray Surveys Simbol. X, May 15, 2007

Local Black Holes and AGN relics Marconi et al. 2004 have shown that local BHs are relics of AGN activity by comparing: the local BH mass function (from galaxy L/σ functions and MBHLbul/MBH-σe) the relic BHMF (from AGN luminosity function and continuity equation) Hard-X LF Soft-X LF Qso LF Importance of AGN LF: even the hard (2 -10 ke. V) XLF does not sample the whole AGN population. Heavily obscured Comptonthick AGN are missing. . . Simbol. X, May 15, 2007 19

G Correction for Compton-Thick sources from XRB models whole AGN pop considered G The only free parameters are the accretion efficiency and Eddington ratio L = ε d. M/dt c 2 L = λ LEdd Simbol. X, May 15, 2007 20

Radiative Efficiency and Fraction of Eddington luminosity G Determine locus in ελ plane where there is the best match between local and relic BHMF! G ε=0. 04 -0. 10 λ=0. 08 -0. 5 which are consistent with common ‘beliefs’ on AGNs CT space density degenerates with accretion efficiency and geometry of reflecting thick matter … 2% 10% 0. 5% Effect of changing scattering efficiency to compute pure reflection spectra of Compton. Thick sources Simbol. X, May 15, 2007 21

Obscured AGN and SMBH census: observation strategies 1. 2. 3. 4. A spectral survey of local CT Seyfert 2 galaxies previously discovered by Beppo. SAX, INTEGRAL Swift and Suzaku. This should be accompanied by a volume limited survey of IR-bright or OIII-bright or … galaxies which have not shown strong X-ray emission below 10 ke. V, to search for highly CT objects. A spectral survey of moderately obscured luminous QSOs discovered by XMM, Chandra up to z~1 (High luminosity) Deep observations to search for faint mildly CT AGN + serendipitous survey (high redshift) A survey of candidate CT AGN selected using their infrared emission (Spitzer/Herschel). Simbol. X, May 15, 2007 22

Summary A large fraction of the 10 -40 ke. V XRB (2/3 - 3/4) will be resolved at a limiting flux of ~ 6 x 10 -15 cgs A major breaktrough is foreseen (compare with ROSAT and Chandra better than ASCA and Beppo. SAX) A large population of CT AGN (as large as Cthin ? ) is waiting to be discovered by sensitive hard ( > 10 ke. V) Xray observations Statistically meaningful sample of the most obscured AGN over a broad range of fluxes and luminosities -> XLF and evolution, SMBH assembly and distribution, geometry of obscuring/reflecting gas, … Simbol. X, May 15, 2007 23