310 likes | 516 Views
Boston, November 2006 Extragalactic X-ray surveys. Spectral analysis of X-ray sources in the CDFS. Paolo Tozzi. 0.3-1 keV 1-3 keV 3-7 keV. Detailed spectral analysis of X-ray sources Tozzi, Gilli, Mainieri et al. 2006, A&A, 451, 457. Rosati et al. 2002 National Geographic, Dec 2002.
E N D
Boston, November 2006 Extragalactic X-ray surveys Spectral analysis of X-ray sources in the CDFS Paolo Tozzi
0.3-1 keV 1-3 keV 3-7 keV Detailed spectral analysis of X-ray sources Tozzi, Gilli, Mainieri et al. 2006, A&A, 451, 457 Rosati et al. 2002 National Geographic, Dec 2002
Deep X-ray Surveys: some open issues The distribution of intrinsic absorption among AGN Detection of Compton-Thick AGN The unresolved fraction of the XRB at high energies
Detected fraction as a function of NH and z z < 0.7 0.7 < z <1.5 z > 1.5 Sampling different luminosity and spectral population at different z Selection effects are relatively easy to model in the X-ray band
NH histogram corrected for completeness Corrected for completeness and sources outside the detectability region <log(NH)> ~ 23.1 dispersion = 1.1 5% C-thick (14 reflection +4 sources NH>1.5 1024 cm-2) density ~ 200 +- 50 deg-2 in agr w XRB model by Gilli, Comastri, Hasinger (2006)
Whole sample (321) QSOII (44) Whole sample (321) Cthick candidates (14) Consistent with evolutionary sequence: pre-QSO phase C-thin absorbed QSO (QSOII @ high z) unobscured QSO activity quiescent spheroidal galaxy Alexander et al. 2005; Stevens et al. 2005 Model: Granato et al. 2005 Part of the missing XRB is from intermediate z strongly absorbed moderate luminosity, possibly C-thick sources, in a secondary, relatively low-z phase of accretion (see “downsizing” or anti- hierarchical behaviour)
The unresolved fraction increases with the energy band ~50% not resolved yet for E> 5 keV (Worsley et al. 2004, 2005) Missing XRB: NH=4.5 1023 cm-2 @ z=0.8
Compute the contribution of the absorbed sources to the XRB NH > 1022 cm-2 NH > 1023 cm-2
Worsley et al. 2004 This work After computing the skycoverage according to the spectral shape of each source
CONCLUSIONS • Detailed X-ray Spectral Analysis of faint X-ray sources: • Towards an universal distribution of intrinsic absorption (selection effects can be easily modelled in X-rays) • >~80% of the AGNs agree with simple unification models (TypeI/TypeII Unobscured/Obscured corr.) • Evidence for strongly absorbed, C-thick sources @ z~1, and a substantial QSOII population at z>~2 • Part of the missing XRB is already observed in agreement with most recent XRB synthesis models
Use secure spectral identifications in CDFS and CDFN 29 galaxies with good spectra in the CDFS and emission line ratios consistent with starbursts or normal galaxies give the X-ray priors. A Bayesian approach allows us to identify 74 galaxies in the CDFS and 136 in the CDFN (2 Ms) Norman et al. 2004
SFR densities XLF consistent with a PLE ~ (1+z)2.7 Consistent with an evolution of SFR (1+z)2.7 for 0<z<1. Compilation from Tresse et al. 2002: Gallego et al. 1995 (H ) Gronwall 1999 Hopkins et al. 2000 Pascual et al. 2001 Tresse et al. 2002 Sullivan et al. 2000 Lilly et al. 1996 Lines from 60 m Saunders et al. 1990, Takeuchi et al. 2003 Norman et al. 2004 XLF of Star Forming Galaxies is a goal for future X-ray missions (Con-X, XEUS)
K20 survey Daddi et al. 2004
IR selected galaxies at z~2 with massive SF Soft Hard Stacked image of 23 BzK galaxies; HR< -0.5 @ 2sigma ; L2-10~1042 erg/sec SFR ~ 170 M⊙ yr-1 (4 higher than LBG). SFRD of 0.04 M⊙ /yr/Mpc3 We are witnessing the massive spheroid formation epoch (the peak of just the low-z tail?) Daddi et al. 2004
The evolution of the obscured fraction with redshift This work Ueda et al. 2003
Average NH artificially increases towards high-z NH ~ 1020cm-2 150 and 80 net counts Tozzi, Gili, Mainieri et al. 2006
Luminosity dependent density evolution: downsizing or anti-hierarchical behaviour Ueda et al 2003 Hasinger et al. 2005
Extended CDFS PI N. Brandt ~1000 sources (Lehmer et al. 2005) 1Ms + 4 X 240 ks
AGN Contribution to the hard XRB 90% resolved in 0.5-2 keV 93% resolved in 2-8 keV AGN contribution 83% in 0.5-2 keV 95% in 2-8 keV (Bauer et al. 2004) CDFS (1Ms): XRB(S> 4.5×10 -16) = (1.70±0.15)×10 -11 erg s -1cm -2deg -2 CDFN (2Ms): XRB(S> 2 ×10 -16) = (2.07±0.15)×10 -11 erg s -1cm -2deg -2
The unresolved fraction increases with the energy band ~50% not resolved yet for E> 5 keV (Worsley et al. 2004, 2005) Missing XRB: NH=4.5 1023 cm-2 @ z=0.8 Worsley et al. 2004; 2005
Worsley et al. 2004 This work After computing the skycoverage according to the spectral shape of each source