Franz. Bauer (IoA -> Columbia)

1
The Evolving AGN Population in the Chandra Deep
Fields
CDF-North 1.945 Ms 450 sq. arcmin 503 sources
CDF-South 0.939 Ms 390 sq. acrmin 326 sources
Franz. Bauer (IoA -gt Columbia) D. Alexander
(IoA), N. Brandt (PSU), A. Fabian (IoA), M.
Worsley (IoA) The CDF/GOODS AGN Teams
2
X-ray Background
(Comastri et al. 1995)
Chandra has resolved 80-95 of the 0.5-2.0 keV
and 70-90 of 2-8 keV cosmic X-ray backgrounds
3
Resolved fraction (Worsley et al. 2004,
submitted)
4
Unresolved fraction (Worsley et al. 2004,
submitted)
(Comastri et al. 1995)
5
Magnitude vs Redshift
525/829 sources in CDFs with redshifts (184 are
phot-z) (Barger et al. 2003, Szokoly 2004,
Mobasher et al. 2003)
Most X-ray sources have optical mags consistent
with L galaxy tracks rather than an MB-23 QSO
Host Galaxy dominates optical light
Can use magnitudes to estimate redshift from
template SED OK for ensemble
6
X-ray/Optical Diagnostic
Fx/Fo is an excellent, relatively
redshift-independent method of crudely
identifying different source types. (e.g.,
Maccacaro et al. 1988 Stocke et al. 1991)
7
X-ray/Optical Diagnostic
X-ray bright sources are comprised mainly of
moderate redshift AGN Whereas X-ray faint
sources are likely to be 1) extension of X-ray
bright AGN to higher redshifts, (e.g.,
Alexander et al. 2001) 2) star-forming
galaxies. (Hornschemeier et al. 2001, 2002,
2003 Alexander et al.
2002 Bauer et al. 2002)
Fx/Fo is an excellent, relatively
redshift-independent method of crudely
identifying different source types. (e.g.,
Maccacaro et al. 1988 Stocke et al. 1991)
8
X-ray/Optical Diagnostic by LX
We can gain some further insight to the number
counts and resolved XRB by making cuts on Fx/Fo
to represent different X-ray source populations.
9
X-ray Spectral Analysis
10
Simple Absorption in CDF AGN (Bauer et al. 2004,
in prep.)
Intrinsic NH Spectral fitting errors do not
change this result. Higher assumed redshifts
could shift things toward more absorption.
11
Simple Absorption in CDF AGN (Bauer et al. 2004,
in prep.)
Some Examples
Soft-Excess BLAGN
Obscured AGN
QSO
NH given in units of 1022 cm-2
12
AGN Demographics
13
AGN Space Density

• Lower luminosity AGN peak at lower redshift
• -gt cosmic down-sizing
• BL AGN (Hasinger et al. 2005, in prep)
• All AGN (Ueda et al. 2003)

14
Optical Host Morphologies
15
Morphologies of AGNs
CDF-North
CDF-South
GOODS ACS (HST) observations provide the deepest
and highest-quality optical data, allowing
unprecedented morphologies
16
Morphologies of AGNs
CDF-North
CDF-South
17
0.4ltzlt1.3 Host Galaxies
(Bauer et al. 2004, in prep.)
18
More Detailed Host Galaxy Constraints
(N. Grogin et al. 2004, in press)
Using the CAS morphological parameterization for
MBlt-19.5, it appears that AGN hosts are
preferentially associated with more concentrated
galaxies (implying they may more bulge-dominated)
compared to field population out to z 1.3,
but show no difference in asymmetry.
19
Morphological Breakdown
(Simmons et al. 2004, in prep.)
20
Future Prospects
21
Still to ComeObservationally

• Deep SIRTF observations (more AGN? mid-IR
  constraints)
• Deep Radio observations (CDF-S deeper CDF-N)
• Larger SCUBA coverage (CDF-S expanded CDF-N)
• Deep near-IR (CDF-N expanded CDF-S)
• Deeper Chandra??? (Compton-thick AGN? better
  spectra)

The CDFs should remain at the observational
forefront in our understanding of faint X-ray AGN
22
Still to ComeScientifically

• More detailed cosmic accretion history which
  hopefully
• includes missing obscured AGN (must find
  first)
• Nature of SMBH/bulge formation and growth (need
  
• measure of bulge out to higher redshifts)

23
Some Conclusions

• Most of the 0.5-8.0 keV background now resolved
  (but not all!)
• Broad variety of source types are detected
• AGN source density (6,000 deg-2) gt15 times that
  in optical
• but few Compton thick AGNs are detected (further
  AGNs to be found)
• Significant fraction of XRB contributors are AGN
  in massive, late-type spiral galaxies -gt
  analogous to local, powerful Seyferts
• Luminous AGN reside in more concentrated
  galaxies
• Bulges in place by z 1.3?
• Not strongly fed by merger activity?

For all papers and data products (CDF-N and
CDF-S) http//www.astro.psu.edu/user/niel/hdf-cha
ndra.html
24
(No Transcript)
25
Trends among these AGN
Colors and CAS parameters for GOODS-N, zlt1.3,
LXgt1042 sources are consistent with spirals as
well
26
X-ray/Optical Diagnostic by LX
Using known (or estimated) redshifts and X-ray
slopes (i.e. NH), we can determine intrinsic,
rest-frame LX
FX/FO tracks LX
27
X-ray Emission Lines in CDF AGN (Bauer et al., in
prep.)
10 of the sample display obvious Fe Ka
emission-line features with EWs from 0.1-1.3
keV. EW0.1 is detection limit stacking can
probe weak lines Only 5-10 potential Compton
thick AGN large
EWs and Glt1.0 characteristic of pure
reflection. (e.g., Maiolino et al. 1998 Bassani
et al. 1999 Matt et al. 2000)
28
CDF Number Counts
Fluctuation analysis (Miyaji Griffiths 2002)
Fluctuation analysis (Miyaji Griffiths 2002)
Bulk of the X-ray bkgd produced by sources near
the knee in the number counts distribution
29
CDF Number Counts by FX/FO
(Bauer et al. 2004, in prep.)
30
CDF Number Counts by LX
(Bauer et al. 2004, in prep.)
0.5-2.0 (2-8) keV XRB dominated by AGN with
log(LX) 44.0 (43.0) log(LX) lt 41.5 sources
(galaxies) match up well with radio estimates
31
CDF Number Counts by NH
(Bauer et al. 2004, in prep.)
0.5-2.0 (2-8) keV XRB dominated by AGN with
log(NH) 21.5 (22.5) This trend illustrates the
basic prediction from XRB synthesis models
32
median props G1.69 NH2x1021 cm-2
Observed value for nearby AGN is G 1.8 before
correcting for effects of reflection or
absorption (e.g., Nandra Pounds 1994)
33
Formation of the Bulge
34
In terms of X-ray observations Deeper vs. Wider?

• Why Go Deeper?
• Discovery space (still approx. photon limited)
• Detect more Compton-thick AGNs
• Improve X-ray spectral analysis
• Detect more galaxies

• Why Go Wider?
• Detect rarer source types (e.g., obscured QSOs,
  high-z AGNs)
• Improve statistics on AGN evolution/luminosity
  function
• Trace both obscured and unobscured AGN evolution
• Uncover extent of large-scale structure (i.e.,
  redshift peaks)

35
X-ray surveys in the 0.5-2.0 keV band
50 times deeper than deepest ROSAT survey
Deep enough to detect mod.lum starbursts at z1
and mod.lum AGNs at z6