The Buildup of Quasars Gordon Richards Drexel University

The Build-up of Quasars Gordon Richards Drexel University With thanks to Michael Strauss, Yue Shen (Princeton), Don Schneider, Nic Ross (Penn State), Adam Myers (Illinois), Phil Hopkins (Berkeley), and a host of other people from the SDSS Collaboration

Caveats I tend to be biased towards: • High redshift (z>1) • High luminosity • Optical Selection • “Quasar” mode accretion • Unobscured (i. e. , type 1) • Quasar=QSO=AGN=any actively accreting supermassive black hole

The Redshift Desert • Redshift desert for galaxies due to lack of spectral features in the optical octave at z~2. • No redshift desert for quasars (in the galaxy sense), but there is in reality. And just as frustrating. Galaxy Franx 2003 Quasar

The z~2. 7 Quasar Desert Observed Corrected Schneider et al. 2007

Z~2. 7 Quasar Colors At 2. 5<z<3. 0 quasars cross through the locus of stars, making those quasars harder to identify (efficiently).

X-ray and IR Selection • X-ray and IR selection don’t suffer from the same problem (and they allow selection of obscured quasars). • But they do have their own problems. • Area surveyed by X-ray is tiny. • Mid-IR has its own 3. 5<z<5 desert. • Not clear that optical/radio/MIR/X-ray selecting same objects (at least at lower luminosity), see Hickox et al. 2008.

Quasar Luminosity Function As with star formation rate, quasars peaked at redshift 2 -3. The rise and fall is even more dramatic in time than redshift. Richards et al. 2006

The Rise of Quasars at z~6 Mere existence z~6 quasars constrains formation models Eddington argument: If the luminosity of a quasar is high enough, then radiation pressure from electron scattering will prevent further gravitational infall. Sets an upper limit to the luminosity for a given mass, or equivalently a minimum mass for a given luminosity. LE = 1. 38 x 1038 M/Msun erg/s ME = 8 x 107 L 46 Msun

Making SMBHs at z~6 The luminosities of the z~6 quasars imply BH masses in excess of 109 MSun. But z~6 is <1 Gyr after the Big Bang. Assembling that much mass in so little time is difficult (but not impossible). Tanaka & Haiman 2009

Quasar Luminosity Function SDSS is relatively shallow. It probes only the tip of the iceberg. Need fainter surveys to get full picture. e. g. , Richards et al. 2006

Cosmic Downsizing Hasinger et al. 2005 Ueda et al. 2003 X-ray surveys probe much deeper. Here we see that peak depends on the luminosity.

Cosmic Downsizing Hasinger et al. 2005 X-ray surveys probe much larger dynamic range. SDSS+2 SLAQ Croom et al. 2009

How does the quasar luminosity function relate to the physics of BH accretion and galaxy evolution?

Quasar Luminosity Function Space density of quasars as a function of redshift and luminosity Typically fit by double power-law Croom et al. 2004

Density Evolution Number of quasars is changing as a function of time.

Luminosity Evolution Space density of quasars is constant. Brightness of individual (long-lived) quasars is changing.

Luminosity vs. Redshift 2. 5 1. 5 4. 5 0. 5 3. 5 Usually we split into L or z instead of making a 3 -D plot, but the information is the same.

Luminosity Evolution Cosmic Downsizing • Pure density or pure luminosity evolution don’t lead to cosmic downsizing. • The slopes must evolve with redshift.

Luminosity Dependent Density Evolution To get cosmic downsizing, the number of quasar must change as a function of time, as a function of luminosity. i. e. , the slopes must evolve.

Bolometric QLF Hopkins, Richards, & Hernquist 2007

Hopkins et al. 2005 Most QLF models assume they are either “on” or “off” and that there is a mass/luminosity hierarchy. Hopkins et al. : quasar phase is episodic with a much smaller range of mass than previously thought. QLF is the convolution of the formation rate and the lifetime. Hopkins et al. 2006

QSO QLF != Galaxy QLF Benson et al. 2003

Hopkins et al. 2005 Most QLF models assume they are either “on” or “off” and that there is a mass/luminosity hierarchy. Hopkins et al. : quasar phase is episodic and “all quasars are created equal” (with regard to mass/peak luminosity). QLF is the convolution of the formation rate and the lifetime. Hopkins et al. 2006

Merger Scenario w/ Feedback • merge gas-rich galaxies • form buried quasars • feedback expels the gas • revealing the quasar • shutting down accretion and star formation e. g. , Kauffmann & Haehnelt 2000 Granato et al. 2004, Di. Matteo et al. 2005, Springel et al. 2005, Hopkins et al. 2005/6 a-z

How Can We Test This? In addition to the evolution of the QLF slopes, we can probe: • The Quasar Luminosity Function • active lifetime (e. g. , Martini 2004) • accretion rate (e. g. , Kollmeier et al. 2006) • MBH distribution (e. g. , Vestergaard & Osmer 2009) • Quasar Clustering • L, z dependence (e. g. , Lidz et al. 2006 ; Shen et al. 2009) • small scales (e. g. , Hennawi et al. 2006; Myers et al. 2008)

Clustering • Red Points are, on average, randomly distributed, black points are clustered • Red points: ( )=0 • Black points: ( )>0 • Can vary as a function of, e. g. , angular distance, (blue circles) • Red: ( )=0 on all scales • Black: ( ) is larger on smaller scales A. Myers

Quasar Clustering • Quasars are more clustered on small scales than large scales. • Comparing with models of dark matter clustering gives the “bias” (overdensity of galaxies to DM) • Linear bias (b. Q=1) ruled out at high significance. Myers et al. 2007

Galaxy Clustering The comoving clustering length of luminous galaxies is roughly independent of z at least to z ~ 5. Therefore, the distribution of galaxies must be increasingly biased relative to the dark matter at high redshift, galaxies=b dark matter Ouchi et al. 2004

How about quasars? Quasars are powered by the ubiquitous super-massive black holes in the cores of ordinary massive galaxies Therefore, we’d expect that the clustering of quasars should be similar to that of luminous galaxies, at the same redshift. Bahcall, Kirhakos et al.

Comoving Correlation Length SDSS Quasars Ross et al. 2009

Quasar Bias Evolution As with galaxies, constant clustering length means strongly evolving bias. Ross et al. 2009

What happens at higher redshift? • If very massive BHs are associated with very massive DM halos, then high-redshift quasars should sit in very rare, many peaks in the density field. • So we expect high-redshift quasars to be more strongly clustered. For 2. 9 < z < 3. 5: r 0=16. 9± 1. 7 Mpc/h; b~10 For z > 3. 5: r 0=24. 3± 2. 4 b~15 Shen et al. 2007

• Use ellipsoidal collapse model (Sheth, Mo & Tormen, 2001) to turn estimates of b. Q into mass of halos hosting UVX quasars. • Find very little evolution in halo mass with redshift. • Our mean halo mass of ~5 x 1012 h-1 MSolar is halfway between characteristic masses from Croom et al. (2005) and Porciani et al. (2004). • This is comparable to the mass of galaxy groups, supporting the idea that quasars are triggered by

Hierarchical Halo Merging CDM theory tells us the expected space density of halos. Comparing with the observed quasar density allows us to determine the fraction of time a quasar is shining. Time for 2 x Mass Time Mass • Lacey & Cole (1993) • Typical quasar hosts double in mass every Gyr or so • Constancy of quasar host halo mass thus limits quasar lifetime to around 106. 5 to 107. 5 yrs

Clustering’s Luminosity Dependence Lidz et al. 2006 old model new model • Quasars accreting over a wide range of luminosity are driven by a narrow range of black hole masses • M- relation mean a wide range of quasar luminosities will then occupy a narrow range of MDMH

No L Dependence for Quasars galaxies Zehavi et al. 2005 quasars Shen et al. 2007

What Next? Measuring bias of faint high-z quasars will break degeneracies between feedback models. bright quasars (e. g. , SDSS) faint quasars (e. g. , LSST) Richards et al. 2006 Hopkins et al. 2007

What We (Used To) Expect 1. Galaxies (and their DM halos) grow through hierarchical mergers 2. Quasars inhabit rarer high-density peaks 3. If quasars long lived, their BHs grow with cosmic time 4. MBH-σ relation implies that the most luminous quasars are in the most massive halos. 5. More luminous quasars should be more strongly clustered (b/c sample higher mass peaks). 6. QLF from wide range of BH masses (DMH masses) and narrow range of accretion rates.

What We Get 1. Galaxies (and their DM halos) grow through hierarchical mergers 2. Something causes the growth of galaxies and their BHs to terminate even as DM halos continue to grow 3. Quasars always turn on in potential wells of a certain size (at earlier times these correspond to relatively higher density peaks). 4. Quasars turn off on timescales shorter than hierarchical merger times, are always seen in similar mass halos (on average). 5. MBH-σ relation then implies that quasars trace similar mass black holes (on average) 6. Thus little luminosity dependence to quasar clustering (L depends on accretion rate more than mass). 7. Need a wide range of accretion rates for a narrow range of MBH to be consistent with QLF.