Heavily-Obscured Super-Massive Heavily-Obscured Super-Massive Black Holes at Low and High Black Holes at Low and High

RedshiftRedshift

Ezequiel Treister (IfA, Ezequiel Treister (IfA, HawaiiHawaii)

Meg Urry, Priya Natarajan, Carie Cardamone, Kevin Schawinski (Yale), Eric Gawiser (Rutgers), Dave Sanders

(IfA) and the MUSYC collaboration

Credit: ESO/NASA, the AVO project and Paolo Padovani

Active Galactic Nuclei Active Galactic Nuclei (AGN)(AGN)

Urry & Padovani, 1995

Black hole–galaxy Black hole–galaxy connectionconnection

All (massive) galaxies have All (massive) galaxies have black holesblack holes

Tight correlation of MBH with Common BH/SFR EvolutionAGN feedback important

All (Massive) GalaxiesAll (Massive) Galaxieshave super-massive black holeshave super-massive black holes

Black hole–galaxy Black hole–galaxy connectionconnection

All (massive) galaxies have black All (massive) galaxies have black holesholes

Tight correlation of MTight correlation of MBHBH with with Common BH/SFR EvolutionAGN feedback important

MMBHBH- - Correlation Correlation

Greene & Ho, 2006

Same relation for bothactive and non-active galaxies.

Black hole–galaxy Black hole–galaxy connectionconnection

All (massive) galaxies have black All (massive) galaxies have black holesholes

Tight correlation of MTight correlation of MBHBH with with Common BH/SFR EvolutionCommon BH/SFR EvolutionAGN feedback important

Common BH/Star Formation Common BH/Star Formation EvolutionEvolution

Marconi et al. 2004

Black hole–galaxy Black hole–galaxy connectionconnection

All (massive) galaxies have black All (massive) galaxies have black holesholes

Tight correlation of MTight correlation of MBHBH with with Common BH/SFR EvolutionAGN feedback importantAGN feedback important

AGN FeedbackAGN Feedback

Springel et al. 2005

No AGN

With AGN Feedback

Supermassive Black Supermassive Black Holes Holes

Credit: ESO/NASA, the AVO project and Paolo Padovani

Many obscured by gas and dustMany obscured by gas and dust

How do we know that?How do we know that?

Local AGN Unification

Explain Extragalactic X-ray “Background”

Observed X-ray Observed X-ray “Background”“Background”

Frontera et al. (2006)

AGN in X-raysAGN in X-rays

Increasing NH

Photoelectric absorptionaffect mostly low energy emission making the observed spectrum look harder.

X-ray BackgroundX-ray Background

Gilli et al., 2007

XRB well explained using a combination of obscured and unobscured AGN.

•Setti & Woltjer 1989•Madau et al. 1994•Comastri et al. 1995•Gilli et al. 1999,2001•Treister & Urry 2005•Gilli et al. 2007•And others…

Compton Thick AGNCompton Thick AGN Defined as obscured sources with

NH>1024 cm-2. Very hard to find (even in X-

rays). Observed locally and needed to

explain the X-ray background. Number density highly uncertain. May contribute significantly to

SMBH accretion. Multiwavelength observations are

required to find them. Hard X-rays (E>10keV) very useful

locally.

SwiftSwift INTEGRALINTEGRAL

ISDC

Swift Sources

Tueller et al. 2007

Significance Image, 20-50 keV

Deep INTEGRAL Survey (3 Msec)Deep INTEGRAL Survey (3 Msec)

Log N-Log SLog N-Log S

Treister et al. 2009

Log N-Log SLog N-Log S

Treister et al. 2009

Fraction of CT AGNFraction of CT AGN

Treister et al. 2009

X-ray background does not constrain density of CT AGN

X-Ray Background X-Ray Background SynthesisSynthesis

Treister et al. 2009

Contribution of CT AGN to the Contribution of CT AGN to the XRBXRB

Only 1% of the XRB comes from CT AGN at z≥2. We can increase the # of CT AGN by ~10x and still fit the XRB.

Only 1% of the XRB comes from CT AGN at z≥2. We can increase the # of CT AGN by ~10x and still fit the XRB.

Treister et al. 2009

CT AGN at High RedshiftCT AGN at High Redshift

Treister et al. 2009

NuSTARNuSTAR

How to find high-z CT AGN NOW?How to find high-z CT AGN NOW?

X-raysX-rays

Tozzi et al. 2006

Trace rest-frame higher energies at higher redshifts Less affected by obscurationTozzi et al. claimed

to have found 14 CT AGN (reflection dominated) candidates in the CDFS.Polletta et al. (2006) report 5 CT QSOs (transmission dominated) in the SWIRE survey.

Fiore et al. 2008

How to find high-z CT AGN NOW?How to find high-z CT AGN NOW?

Mid-IRMid-IR X-ray StackingX-ray Stacking

FF2424/F/FRR>1000>1000

FF2424/F/FRR<200<200

• 4 detection in X-ray stack. Hard spectral shape, harder than X-ray detected sources.Good CT AGN candidates.• Similar results found by Daddi et al. (2007)

Extended Chandra Deep Field-Extended Chandra Deep Field-SouthSouth

Area:Area: 0.3 deg2

X-rays:X-rays: Chandra 250ks/pointing

Optical:Optical: Broad band UBVRIz (V=26.5)+ 18 Medium band filters (to R=26)

Near-IR:Near-IR: JHK to K=20 (Vega)

Mid-IR:Mid-IR: IRAC 3.6-8 microns + MIPS 24 microns to 35 µJy

Spectroscopy:Spectroscopy: VLT/VIMOS, Magellan/IMACS (optical) VLT/SINFONI, Subaru/MOIRCS (near-IR)

Mid-IR SelectionMid-IR Selection

R-K (Vega)

- 211 sources with f24m/fR>1000 and R-K>4.5- f24m>35Jy- 18 X-ray detected

Treister et al. ApJ in press

Redshift DistributionRedshift Distribution

All Sources

X-ray Detected

X-ray Undetected- Photo-z for ~50% of the sources- X-ray sources brighter at all wavelengths- Spec-z for 3 X-ray sources and photo-z for 12 (83% complete).

Treister et al. ApJ in press

Hardness Ratio Hardness Ratio N NHH

X-ray sources with redshift onlyX-ray sources with redshift only

- 2 unobscured- 11 obscured Compton-thin- 2 Compton Thick

Assumed fixed =1.9

Treister et al. ApJ in press

Stacking of non-Xrays SourcesStacking of non-Xrays Sources

Soft (0.5-2 keV) Hard (2-8 keV)

- ~4 detection in each band.- fsoft=2.1x10-17erg cm-2s-1. Fhard= 8x10-17erg cm-2s-1

- Sources can be detected individually in ~10 Msec.- Hardness ratio 0.13, NH=1.8x1023cm-2.- Alternatively, ~90% CT AGN and 10% star-forming galaxies.- Some evidence for a flux dependence. >95% CT AGN at the brightest bin, 80% at the lowest. Large error bars.

Treister et al. ApJ in press

Rest-Frame StackingRest-Frame Stacking

Good fit with either NH1023cm-2 or combination of CT AGN with star-forming galaxies.

Consistent results with observed-frame stacking.

Treister et al. ApJ in press

X-Ray to Mid-IR RatioX-Ray to Mid-IR Ratio

Both X-rays and 12µm good tracers of AGN activity.Observed ratios for X-ray sources consistent with local AGN (dashed line).

Treister et al. ApJ in press

X-Ray to Mid-IR RatioX-Ray to Mid-IR Ratio

Effects of obscuration in X-ray band luminosity.

Only important for Compton Thick sources.

Treister et al. ApJ in press

X-Ray to Mid-IR RatioX-Ray to Mid-IR Ratio

~100x lower ratio for X-ray undetected sources.

Explained by NH~5x1024 to 1025cm-2

Treister et al. ApJ in press

X-Ray to Mid-IR RatioX-Ray to Mid-IR Ratio

Ratio for sources with L12µm>1043erg/s (~80% of the sources) ~2-3x higher than star-forming galaxies

Treister et al. ApJ in press

X-Ray to Mid-IR RatioX-Ray to Mid-IR RatioUsing Lx/L12µm=0.007 to separate AGN and star-forming galaxies ~80% AGN, consistent with HR value.

In sources with L12µm>1044erg/s outside selection region fraction of AGN ~10%.

Treister et al. ApJ in press

Near to Mid-IR ColorsNear to Mid-IR Colors- Distributions significantly different- X-ray detected sources much bluer- Average f8/f24=0.2 for X-ray sources and 0.04 for X-ray undetected sample

BluerRedder

Well explained by different viewing angle (30o vs 90o) in the same torus model Can it be star-formation versus AGN?

Treister et al. ApJ in press

Near to Mid-IR ColorsNear to Mid-IR Colors

Armus et al. 2007

ULIRG, LINER

ULIRG, Sey2

MorphologiesMorphologies

Ground based: K,H,R

HST/WFC3J,H,Y

GOODS-SUDF

Optical/Near-IR SED FittingOptical/Near-IR SED Fitting

X-ray Detected

X-ray Undetected

X-ray Detected

X-ray Undetected

- Median stellar mass for X-ray detected sources ~4.6x1011 Msun.- For X-ray undetected source ~1011 Msun.

- Mild extinction values found in general.- Maximum Av~4 mags.- Median E(B-V)=0.6 for X-ray detected sources and 0.4 for undetected ones.

Treister et al. ApJ in press

Evidence for significant recent star formation in most sources

Heavily-Obscured AGN Space Heavily-Obscured AGN Space DensityDensity

Systematic excess for Lx>1044erg/s sources relative to extrapolation of Compton-thin LF

Strong evolution in number of sources from z=1.5 to 2.5.

Consistent with heavily-obscured phase after merger?

Treister et al. ApJ in press

Obscured to Unobscured Quasar Obscured to Unobscured Quasar RatioRatio

Treister et al. in prep.

The Merger-Quasar ConnectionThe Merger-Quasar Connection

Treister et al. in prep.

Obscured quasars are the product of the merger of two massive gas-rich galaxies. After a time t the quasar becomes unobscured.

€

NobscNUnobsc

(z) =Δtd2merger

dtdNNgal (> Mmin (z)) fgas(z)

NUnobsc (z)

The Merger-Quasar ConnectionThe Merger-Quasar Connection

Treister et al. in prep.

t=7716 Myrs

SummarySummary

• Number of mildly CT sources at z~0 lower than expected.

• Only ~1% of XRB comes from CT AGN at z~2.• Multiwalength surveys critical to find high-z CT AGN.

• Mid-IR selection finds large number of CT AGN at z>1.5.

• Morphologies indicates interactions/mergers.

• Host galaxy masses ~1011 Msun with young and obscured stellar populations.

• Strong evolution in numbers up to z~3.• This could be evidence for a heavily obscured phase after quasar triggering.

• Number of mildly CT sources at z~0 lower than expected.

• Only ~1% of XRB comes from CT AGN at z~2.• Multiwalength surveys critical to find high-z CT AGN.

• Mid-IR selection finds large number of CT AGN at z>1.5.

• Morphologies indicates interactions/mergers.

• Host galaxy masses ~1011 Msun with young and obscured stellar populations.

• Strong evolution in numbers up to z~3.• This could be evidence for a heavily obscured phase after quasar triggering.