The build-up of SMBHs in z<1 red galaxies

1
The build-up of SMBHs in zlt1 red galaxies

• Kate Brand, STScI
• Galaxy and BH evolution Towards a unified view
• Tucson, 29th Sept 2007
• Collaborators Michael Brown, Richard Cool,
  Vandana Desai, Arjun Dey, Buell Jannuzi, Emeric
  Le Floch, John Moustakas, Tom Soifer the
  Xbootes, IRAC and AGES teams.

2
Questions

• How did the mass of SMBHs grow?
• Have SMBHs accreted a significant fraction of
  their mass between z1 and the present?
• How is this related to the build-up of the red
  sequence?
• What is connection between SMBH growth and
  massive galaxy evolution?

3
Outline

• A sample of red galaxies from the NDWFS Bootes
  field
• X-ray 24?m stacking
• Optical line diagnostics
• Infrared color-color diagram
• The build up of SMBHs between z1 and z0

4
The NDWFS Bootes field
9 deg2 Bw, R, I, K 27.1, 26.1, 25.4, 19.0 mag
(Vega).
PIs A. Dey B. Jannuzi
5
Multi-wavelength observations in the Bootes field
VLA P-band 90 cm 7
sq.deg. 100mJy 100
complete van Breugel, PI VLA L-band 21
cm 1 sq.deg. 15mJy
100 complete Higdon, PI VLA (FIRST)
21 cm 9 sq.deg. 1mJy
100 complete
public Westerbork 21 cm
7 sq.deg. 8mJy 100
complete Rottgering, PI Spitzer/MIPS
24,70,160um 9 sq.deg. 3.0, 30, 100 mJy 100
complete Jan 2004 GTO Spitzer/IRAC 3.6 - 8um
9 sq.deg. 6.4, 8.8, 51, 50mJy 100
complete Eisenhardt et al. Spitzer/IRAC
3.6,4.5,5.8,8um 9 sq.deg. 3.2, 4.4, 25,
25mJy Stern et al. large GO5 Spitzer program
NOAO/FLAMEX J, Ks 4.7
sq.deg. 19.3 mag 100 complete
Elston et al. (2005) NOAO K,
Ks 9 sq.deg. 18.6 mag
100 complete NOAO J,
H 9 sq.deg. 21 mag
40 complete NOAO BW,
R, I 9 sq.deg. 25.5-26.6 mag 100
complete NOAO U
9 sq.deg. 25 AB mag 100
complete GALEX FUV, NUV
1 sq.deg. 26 AB mag 100
complete, GTO GALEX FUV, NUV
9 sq.deg. 25 AB mag in
progress, GTO HST I, H
sparse 26, 23 mag
in progress Chandra 0.5-7 keV
9 sq.deg. 4e-15 erg/s/cm2 100
complete NOAO/Keck spectroscopy
sparse 24 mag in
progress (500 so far) MMT/Hectospec spectra
9 sq.deg. R20.5 mag completed (25,000
redshifts) Spitzer/IRS spectroscopy
sparse
in progress
6
The red galaxy sample

• Reliable photometric redshifts for entire NDWFS
  catalog from ANNZ calibrated using 20,000
  galaxies from AGES (Brown et al. 2005).
• U-V vs. Mv color selection

? 26,000 red galaxies
7
Nuclear accretion in red and dead galaxies at
zlt1

• Used Chandra XBootes survey to stack up the X-ray
  images to obtain a mean X-ray luminosity.
• 5-ks on a single object -gt for 1000 galaxies, a
  5-Ms observation on the mean object.

Brand et al. (2005) Brand et al. in prep.
3200
6400
7900
8900
8
Contribution of LMXBs and HMXBs

• HMXBs
• Short lifetime ? Lx ?SFR
• (Grimm et al.2003)
• - Population Synthesis fits to optical photometry
  ? SFR
• ? Expected Lx

X-ray Luminosity
SFR

• LMXBs
• long lifetime ? Lx ? stellar mass
• (Kimm et al.2004)
• - Population Synthesis fits to optical photometry
  ? abs K-mag
• ? Expected Lx

X-ray Luminosity
Lk
9
Basic X-ray stacking results for red galaxies

• ltLxgt 1041-1042 ergs s-1 is 5-10x too high to be
  due to stellar sources.
• The X-ray emission is dominated by accretion
  onto a SMBH

• ltLxgt is 10-100x fainter than typical Seyferts
• The accretion rate is very low and/or
  radiatively inefficient

• The X-ray spectrum is hard and can be explained
  by
• Absorbed ?1.7 power-law with NH 1 - 5 x 1022
  cm-2 (increasing with z)
• Unabsorbed ?0.7 power-law (e.g. ADAF)
• The AGN must be obscured or radiatively
  inefficient

Brand et al. in prep.
10
The mean X-ray luminosity increases with redshift
Absorbed
Total
Hard
Lx
Soft
Stellar
Unabsorbed
Total
Log (1z)
Hard
Lx ? (1z) 6.6 0.9
Soft
Stellar
Brand et al. in prep.
11
The mean X-ray luminosity is higher for more
massive galaxies
Stellar Mass (x1E10 M0)
Stellar Mass (x1E10 M0)
12
24?m stacking of red galaxies
0.2 lt z lt 0.4 0.4 lt z lt 0.6

• 700/18000 (4) of red galaxies have f24gt0.3 mJy
  (ltf24gt 0.5 mJy)
• 22/49 (45) of X-ray detected sources have
  f24gt0.3mJy.
• Mean 24?m flux is far larger than expected by
  mass loss from evolved stars (cf. Rodighiero et
  al. 2007, Davoodi et al. 2006).

0.6 lt z lt 0.8 0.8 lt z lt 1.0
13
The mean infrared luminosity increases with
redshift
LIR ? (1z) 6.8 1.0
LIR ? (1z) 8.6 1.3
14
The mean infrared luminosity is a less strong
function of mass
Same plots for X-ray luminosity
15
Optical line diagnostics
Yan et al. 2006
Optical spectra from the AGN and Galaxy Evolution
Survey (AGES Kochanek et al. in prep)
16
IRAC color-color diagram
Red galaxies with 24?m emission have redder
5.8-8.0 colors than the general red galaxy
population ? PAH emission ? a tracer of star
formation.
17
How important is the zlt1 low accretion phase in
the build-up of SMBHs?
.
.

• ? M Lbol/c2
• M 1.76 ?10-5 M?/yr (0.1/?) (Lx/1041) x CB
• radiative efficiency of accretion energy
• CBbolometric correction
• MEDD 2.2 (MBH/108) M?/yr
• Eddington ratio M / MEDD

.
.
.
18
How important is the zlt1 low accretion phase in
the build-up of SMBHs?
Integrating accretion rate from z1 to present
(? 0.1) ? increase in BH mass 5 ?107
M? MBH 0.002 Mbulge ? ltMBHgt 1 x 108
M? at z1 ? SMBH mass could have increased by
50 since z1 in this low accretion-level
regime half of this build-up due to 1 of
population (bursty activity) complications -
Contribution from star-formation? - Non-static
population (red sequence doubling between z1 and
the present) - Sensitivity to spectral shape
(absorption correction) - Bolometric luminosity
correction - if use 20, half the inferred
growth - Accretion efficiency ( if ? 0.001
(ADAF) ? increase in BH mass 5 ?109
M? if no outflows)
19
Summary

• We have analyzed the X-ray, optical, and infrared
  light emitted from 20,000 red galaxies in the
  NDWFS Bootes field.
• An X-ray stacking analysis shows that the X-ray
  emission is dominated by AGN activity - low level
  radio mode accretion rates?
• The nuclear accretion rate increases
  significantly with redshift and is larger for
  more massive red galaxies.
• Infrared stacking shows more infrared emission
  than expected for red and dead galaxies. The
  mean infrared luminosity increases with redshift
  but relation with mass is less strong.
• Although X-ray emission is likely to be dominated
  by AGN activity, infrared light may be more
  dominated by star formation activity, and optical
  emission may be a mixture of the two.
• The X-ray analysis shows that the average SMBH
  mass could have increased by 50 between z1 and
  the present.

20
The End
21
The evolving Luminosity function of red galaxies
Luminosity Function
Luminosity density of all red galaxies - 50
increase in luminosity density between z0 and
z1 - Total stellar mass has doubled since z1
Only gt4 L red galaxies - 80 of the stellar
mass in massive galaxies was already in place at
z0.9.
Brown et al. 2006