Xray selected z3 QSOs in the XMMCOSMOS field

1
X-ray selected zgt3 QSOs in the XMM-COSMOS field
Marcella Brusa (MPE)
The XMM-COSMOS and COSMOS teams .. and in
particular A. Comastri, R. Gilli, G.
Hasinger, K. Iwasawa, V. Mainieri, M. Mignoli,
M. Salvato, G. Zamorani A. Bongiorno, N.
Cappelluti, F. Civano, F. Fiore, A. Merloni, J.
Silverman, J. Trump, C. Vignali, P. Capak, M.
Elvis, O. Ilbert, C. Impey, S. Lilly
2
Why high-z AGNs are interesting?

• locally MBH0.0013?MBulge
• SDSS QSO zgt6 gt MBH3-7?109MSun
• (Willott 03)
• ? Either M1012MSun bulges are in place at zgt6
  or the MBH-Bulges-properties relationships must
  break down at high-z
• ?Forming (enough) 109-10MSun BHs and (possibly)
  1012MSun Bulges at zgt6 can be a challenge for
  models of structure formation.
• As well as forming metals and dust.
• Strong constraints on
• cosmological models

adapted from Ferrarese Merritt 2000, Gebhardt
et al. 2000
3
X-rays from High-z QSOs
1990-1994 pioneering works with ROSAT
Wilkes92 Elvis94 Bechtold94 (record QSO
z4) 2002-2005 Chandra/XMM contribution Follow-u
p of optically SDSS QSOs Brandt02 Mathur02
Vignali03,05 (record QSO z6.4) 2015-2020 XEU
S ? record z 10?
Vignali05
4
The population of zgt3 QSO

• Radio QSO
• (Wall et al., 2005)
• Optical QSOs
• (Schmidt95, Fan01,04,
• Richards06)
• Exponential decline in space density
• at z2.7
• X-ray What happens at zgt3?
• decline or costant ?
• Soft X-ray ROSAT/Chandra/ XMM
• (Hasinger, Miyaji Schmidt 2005)
• Chandra/ROSAT
• (Silverman et al. 2005/2008)
• statistics still low at z3-5
• (NO statistics at zgt6)

XMM Chandra ROSAT
?
eROSITA XEUS
5
1608 soft 0.5-2.0 keV 1103 medium 2.0-4.5
keV 250 hard 4.5-10.0 keV
Cosmos Survey 2 deg2 (PI N. Scoville) with
deep multiwavelength coverage at all lambda
XMM-COSMOS survey (PI G. Hasinger) Mosaic of
55 XMM observations, 1.4 Ms 35 ks average One
of the main goal Characterization of The zgt3
QSO population
Hasinger07, Cappelluti07 See poster by Nico
Cappelluti
6
zgt3 QSO counts expectations

• at S(0.5-2 keV) gt1e-15 cgs
• Predictions
• (from XRB models
• Gilli, Comastri Hasinger 07)
• unabsabs
• 30 deg-2 (exponential cutoff at z.2.7)
• Schmidt et al. 1995
• 80 deg-2 (constant evolution
• extrapolating Hasinger05 /
• La Franca05 LF)

Models from Gilli07
7
XMM-COSMOS sources redshifts
compilation from ongoing spectroscopic projects
IMACS/MagellanVLT/ESO SDSS literature
data

• Flux limited sample (50 of the area coverage in
  at least one band) at 10-15 cgs
• 1651 XMM sources
• lt10 problematic ID thanks to
• IRChandra info
• 670 secure spectroscopic redshifts (40)
• Incompleteness especially for high-z and Type 2
  AGN

Blue BL AGN Black ALL AGN
(adapted from Brusa et al. 2007 ApJS)
8
AGN photometric redshifts
Photo-z computed using gt30 bands SDSS, Subaru
including IB, CFHT, J, K, IRAC..

• LESS than 10
• catastrophic outliers
• (to be compared with
• COMBO-17, Wolf et al. 2004)

Salvato et al. (in prep)
9
AGN redshift distribution in XMM-COSMOS
X-ray sample (AGN) Empty speczphotz Filled
specz
zgt3 sample 40 objects (22 specz 18
photoz) Additional 14 objects, no photoz
available. Candidates very high-z AGN (EXOs,
Koekemoer et al. 2004) Brusa et al. ApJ
submitted
10
Optical spectra (from VIMOS and IMACS)
11
Photometric redshifts
12
General properties
Magnitude distribution I20-26 (25 Igt24)
Luminosities distribution Log(Lx) 44 - 45
ACS image, 15x15 I23.5
13
Optical colors
Color color selection v-I vs. b-v (proposed,
e.g. in Casey et al. 2008, Siana et al. 2007)
Lower-z (0-2)
Interm-z (2-3)
High-z (gt3) locus
8 objects would not have been selected
40 (magenta/yellow) contaminants
14
X-ray properties
logNHlt23
HR vs. redshift
logNHgt23
logNHlt23, BL
logNHlt23 NOT BL

• Ratio of obscured/unobscured objects
• is 20

15
Contribution to source counts
Lower bound 22 spectro-z Upper-bound adding 10
EXOs Dashed line Expectations from XRB models
extrapolating Hasinger05 LF Solid
line Exponential decay introduced at z2.7
(Schmidt95) Flat evolution completely ruled
out Tightest constraints to date (largest and
cleanest sample)
Models Gilli, Comastri Hasinger 2007, AA
16
Space densities
Red curve predictions logLxgt44.2 AGN
(unobsobs) Gilli07 using Hasinger05 and La
Franca05
Dashed area (rescaled) space density for
optically selected bright QSO Richards2006,
Fan2001 Blue curve Silverman08 LF, Ilt24
sample
17
Summary

• Results based only on XMM-COSMOS selection and
  exploitation of COSMOS database (no need to
  combine with other X-ray surveys )
• Flat evolution at zgt3 definitively ruled out
  (space densities number counts)
• 32/40 (80) X-ray selected quasars would have
  been selected also by optical color-color
  criteria (but with a comparable number of low-z
  contaminants)
• ?not a significant population with unusual
  optical colors
• Sizable sample of obscured AGN (BAL, NL..)

18
Whats next?

• High-z quasars
• ? eROSITA (high-L)
• 2500 zgt3 QSOs in the Deep survey
• 400 deg2 on northsouth
  pole, F(0.5-2) gt 4e-15 cgs
• 25000 zgt3 QSOs in the All Sky survey
• 30.000 deg2 all sky,
  F(0.5-2) gt 1e-14 cgs
• ? XEUS (low-L and very high-redshift)
• ? joint multiwavelength campaigns
• Synergies with other (big) observatories
  is mandatory
• - PanSTARRS (all sky) GROND (for
  faintest sources)
• - SDSS-like survey? LAMOST ?
• - (ESO facilities can be used on 200
  deg2 South pole)
• - LBT/GTC deep UBband imaging (on 200
  deg2)
• - CFHTLS Wide Survey 100 deg2
  (u,g,r,i,z) AB25-26