Host galaxies and large-scale structures of active galactic nuclei
Ryan C. Hickox, Stephanie M. LaMassa, John D. Silverman, Alexander Kolodzig
AAstronomy in Focus, Volume 1XXIXth IAU General Assembly, August 2015Piero Benvenuti, ed. c (cid:13) Host galaxies and large-scale structures ofactive galactic nuclei
Ryan C. Hickox , Stephanie M. LaMassa , John D. Silverman ,Alexander Kolodzig Department of Physics and Astronomy, Dartmouth College, 6127 Wilder Laboratory,Hanover, NH 03755, USAemail: [email protected] NASA Goddard Space Flight Center, Code 662, Greenbelt, MD 20771, USA Kavli Institute for the Physics and Mathematics of the Universe (WPI), The University ofTokyo Institutes for Advanced Study, The University of Tokyo, Kashiwa, Chiba 277-8583,Japan Kavli Institute for Astronomy and Astrophysics, Peking University, Beijing 100871, China
Abstract.
Our understanding of the cosmic evolution of supermassive black holes (SMBHs) has beenrevolutionized by the advent of large multiwavelength extragalactic surveys, which have enableddetailed statistical studies of the host galaxies and large-scale structures of active galactic nuclei(AGN). We give an overview of some recent results on SMBH evolution, including the connectionbetween AGN activity and star formation in galaxies, the role of galaxy mergers in fueling AGNactivity, the nature of luminous obscured AGN, and the connection between AGN and theirhost dark matter halos. We conclude by looking to the future of large-scale extragalactic X-rayand spectroscopic surveys.
Keywords. (galaxies:) quasars: general, galaxies: Seyfert, surveys, X-rays: galaxies, (cosmol-ogy:) large-scale structure of universe
1. Introduction
The past two decades have seen great progress in understanding the growth and evo-lution of supermassive black holes (SMBHs) over cosmic time (for one of several recentreviews see Alexander & Hickox 2012). It is now well-established that SMBHs obtainthe bulk of their mass through accretion of matter, observable as active galactic nuclei(AGN), and that there are connections between the cosmic growth of SMBHs and thatof their host galaxies, due to common evolution histories or to feedback processes thatlink the growth of SMBHs to the state of gas and star formation in their host systems(see Fabian 2012 and Kormendy & Ho 2013 for recent reviews).Recently, large multiwavelength extragalactic surveys have enabled breakthroughs inunderstanding the AGN-galaxy connection through detailed statistical studies of the hostgalaxies and large-scale structures of AGN. The resulting insights into SMBH evolutionare analogous to the understanding of galaxies that emerged from the large redshift sur-veys in the 2000’s (e.g., Strateva et al. 2001; Zehavi et al. 2005; Blanton 2006; Coil et al.2006). Particularly valuable observational resources have been the
Chandra X-ray Obser-vatory (Tananbaum et al. 2014) and
XMM-Newton (Jansen et al. 2001), for performingdeep X-ray surveys detecting large numbers of AGN to high redshift; the
Herschel SpaceObservatory (Pilbratt et al. 2010) to constrain star formation rates (SRFs) in AGN hostgalaxies; the
Spitzer Space Telescope (Werner et al. 2004) and
Wide-Field Infrared SurveyExplorer ( WISE ; Wright et al. 2010) for identifying large numbers of luminous, obscured1 a r X i v : . [ a s t r o - ph . GA ] D ec Hickox et al.quasars; the
Nuclear Spectroscopic Telescopic Array ( NuSTAR
Harrison et al. 2013) forstudying the high-energy emission from heavily obscured AGN; and extensive follow-upof AGN with ground-based multi-object spectrographs.In this Proceedings, we begin with an overview of recent progress on the connectionbetween AGN activity and star formation (SF) in galaxies. We then discuss the linkbetween AGN activity and galaxy mergers, highlight observational studies of the lumi-nous, obscured AGN that may represent an important phase in the evolution of massivegalaxies, and discuss the utility of AGN clustering measurements in understanding theconnection between AGN and their host dark matter (DM) halos. Finally, we look to-ward the future of statistical studies of AGN host galaxies and structures with the nextgeneration of very large X-ray and spectroscopic surveys.
2. The AGN star-formation connection
There is compelling indirect evidence for a global connection between AGN activ-ity and SF in galaxies, from the tight correlation between SMBH masses and galaxyproperties (e.g., McConnell & Ma 2013; Kormendy & Ho 2013) and the similar cosmicevolutionary histories of these two processes (e.g., Merloni & Heinz 2013; Kormendy &Ho 2013). We can now probe this connection directly by observing the SF properties ofAGN host galaxies. One useful observational tool is the distribution of galaxy colors andluminosities (e.g., Strateva et al. 2001; Blanton 2006), which clearly separates galaxiesinto two populations: blue, star-forming, relatively low-mass galaxies, and red, passive,higher-mass systems. When we locate AGN host galaxies (out to redshifts z ∼
1) incolor-luminosity space, we find that the hosts of radiatively efficient, rapidly growingSMBHs (identified as AGN based on X-ray, infrared, or optical line emission) are pre-dominantly located among the star-forming systems (e.g., Nandra et al. 2007; Hickoxet al. 2009; Goulding et al. 2014; Mendez et al. 2015). There is a preference for moreluminous AGN to be found in the more massive, redder end of the star-forming galaxypopulation (e.g., Schawinski et al. 2009, 2010b). perhaps connected to the higher Ed-dington limit associated with their more massive SMBHs (e.g., Aird et al. 2012; Trumpet al. 2015).We can further study the host galaxies of mechanically -dominated AGN identified byradio synchrotron emission from relativistic jets. In contrast to the radiatively-efficientAGN, radio AGN are found predominantly in massive, passive galaxies, generally avoidingthe star-forming systems (e.g., Hickox et al. 2009; Goulding et al. 2014; Mendez et al.2015). (Note that AGN with radio jets and high-excitation emission lines are generallyfound among the star-forming galaxies; Smolˇci´c 2009.) These results point toward ageneral picture in which rapidly-growing, radiatively-efficient AGN are found in star-forming galaxies, fueled by the supply of cold gas that also produces the SF, whilemechanically-dominated, slowing growing AGN are found in passive galaxies, producingthe feedback required to stop cooling of hot gas in their massive halos (e.g., Bower et al.2006; Croton et al. 2006).Recent studies have built on this work by taking advantage of
Herschel for reliablemeasures of SFR in AGN host galaxies based on far-IR emission, where there is littlecontamination from the AGN (e.g., Mullaney et al. 2011a). These results have shownthat when averaging over the full star-forming galaxy population, the average SMBHgrowth is correlated with SFR (e.g., Mullaney et al. 2012), with a linear relationshipbetween SFR and average accretion rate (e.g., Chen et al. 2013, Figure 1). However, thepicture becomes more complicated when looking at individual
AGN. The colors, SFRs,spatial clustering, and merger rates of AGN hosts are essentially indistinguishable from
M 6. AGN host galaxies and structures L IR [ L O • ])4243444546 l og ( < L AGN > [ e r g s - ]) z = 2.00 z = 1.15 z = 0.65 z = 0.35 z = 0.05Symeonidis et al. (2011)Chen et al. (2013)Rafferty et al. (2011) (a) m o d e l -8 -6 -4 -2 E dd i ng t on r a ti o AGN variability from simulation (Novak et al. 2011)AGN“normal” galaxy
Figure 1.
Left:
The observed correlation between SFR (as measured by far-IR luminosity) instar-forming galaxies and the average AGN luminosity (see Hickox et al. 2014 for references).Colored lines show the linear relationship assumed in the model of (Hickox et al. 2014).
Right:
Illustration of AGN variability. The bottom panel shows the Eddington ratio versus time forthe simulation of Novak et al. (2011). Image credits, from left: M81: Wikimedia Commons; PG0052+251: J. Bahcall (IAS, Princeton), M. Disney (Univ. of Wales), NASA/ESA. typical star-forming galaxies of similar mass (e.g., Cardamone et al. 2010; Xue et al.2010; Mullaney et al. 2011b; Cisternas et al. 2011; Treister et al. 2012; Mendez et al.2015). The average SFR of AGN host galaxies at depends strongly on redshift similarlyto inactive galaxies, but shows little if any dependence on AGN luminosity (e.g., Shaoet al. 2010; Rosario et al. 2012; Stanley et al. 2015).We therefore are presented with a puzzle in which there is a strong global correlation be-tween SF and AGN activity, but the links are weak if non-existent for individual sources.This raises the question: ”why are only a fraction of star-forming galaxies observed asAGN?” The answer appears to lie in the stochastic variability of AGN. Simulations (e.g.,Novak et al. 2011; Gabor & Bournaud 2013) and observations of AGN light echoes (e.g.Schawinski et al. 2010a; Keel et al. 2015) suggest that AGN accretion can vary overmany orders of magnitude on timescales of 1 Myr or less, much shorter than the typicalevolution timescale for galaxies. We might therefore think of all star-forming galaxies ashosting an AGN, when averaged over >
100 Myr timescales. Hickox et al. (2014) pre-sented a simple analytic model for the AGN population based on a direct connectionbetween SMBH accretion and star formation and including stochastic variability. Thissimple model is able to reproduce the observed relationships between AGN luminosity,SFR, and merger rates, as well as the general evolution of the AGN luminosity function.Other recent simulations have found similar results (e.g., Thacker et al. 2014; Volon-teri et al. 2015), highlighting the need for future studies to not simply compare AGNhost galaxies with their inactive counterparts, but to measure the distribution of AGNaccretion rates as a function of host galaxy properties.Given the clear connection between SMBH growth and SF, an important questionarises regarding the nature of AGN feedback , and whether in some cases the AGN canshut down star formation by heating or removing the gas supply (see Fabian 2012 for areview). Some simulations show that outflows from radiatively efficient AGN can have astrong effect on galaxy-scale gas (e.g., Di Matteo et al. 2005; Booth & Schaye 2011), butother models of AGN in disk galaxies show a limited impact of the AGN on the star-forming disk (Gabor & Bournaud 2014). Some theoretical studies also indicate that AGNactivity can trigger
SF through positive feedback (e.g., Zubovas et al. 2013; Nayakshin Hickox et al. m e r g e r fr ac ti on
42 43 44 45 46 47log( L AGN [erg s -1 ]) z = 0.25 z = 1.00 z = 1.75mergers, interactions, and irregularsmergers and interactions inactive systems
10 0 10 20 30 40 50 60 70 800.51.01.52.02.53.03.54.04.5 z, M ,δ z, M ,δ ,M h ,D
10 0 10 20 30 40 50 60 70 800246810121416 A G N e x c e ss ( p a i r s / c o n t r o l ) z, M ,δ z, M ,δ ,M h ,D
10 0 10 20 30 40 50 60 70 80 r p ( h − kpc ) z, M ,δ z, M ,δ ,M h ,D Figure 2.
Left:
Relationship between the fraction of AGN in mergers and AGN luminosity takenfrom the compilation of Treister et al. (2012). The gray shaded area indicates the typical rangeof merger fractions for inactive galaxies in the control samples studied by Cisternas et al. (2011)and Kocevski et al. (2012). The colored curves show the predictions of the Hickox et al. (2014)model assuming a correlation between merger fraction and L IR determined by Kartaltepe et al.(2012). Right:
The fraction of AGN in galaxy pairs and post-mergers (shown in the grey area atleft) relative to matched control galaxies versus projected separation. Optically selected, mid-IRselected and radio selected are shown in the top, middle and bottom panels respectively. Allthree samples show an elevated AGN fraction over the control galaxies; the optical and mid-IRAGN shown an increasing enhancement in AGN activity with decreasing pair separation.
3. Fueling of AGN by galaxy mergers
An important aspect of the connection between SMBHs and galaxies has been therole of galaxy mergers in fueling AGN activity. This has long been a matter of debate,with some studies showing a clear link between AGN activity and mergers (e.g., Bahcallet al. 1997; Urrutia et al. 2008; Koss et al. 2010; Glikman et al. 2015), while other studiesshow effectively no difference in the merger rates between AGN and inactive galaxies (e.g.,Grogin et al. 2005; Gabor et al. 2009; Cisternas et al. 2011; Kocevski et al. 2012; Mechtleyet al. 2015). Treister et al. (2012) suggested that these differences can be reconciled whenconsidering AGN as a function of luminosity, with powerful quasars commonly found inmergers, while less luminous AGN have merger rates of 20% or less, similar to ”normal”galaxies (Figure 2, left ). While this work necessarily relies on heterogeneous definitionsof ”mergers”, the results are broadly suggestive of a trend with AGN luminosity.A powerful way to test the merger-AGN connection is by directly tracing AGN activityalong the merger sequence, using kinematic pairs of galaxies. Kinematic pairs are well-known to show an enhancement of SFR that increases with decreasing separation (e.g.,Ellison et al. 2008; Woods et al. 2010; Kampczyk et al. 2013). The incidence of AGNactivity shows very similar behavior (Figure 2, right ), for AGN identified from opticalemission lines (Ellison et al. 2015), X-ray luminosity (Silverman et al. 2011), or mid-
M 6. AGN host galaxies and structures Figure 3.
Hubble Space Telescope
WFC3 images of red QSOs, described in Glikman et al.(2015), showing the observed images (top) and the residuals after subtracting smooth quasarand galaxy models (bottom). The majority of the red QSOs show disturbances characteristic ofmajor mergers. (Image credit: E. Glikman/NASA) infrared (IR) colors (Satyapal et al. 2014). Radio-loud AGN with low-excitation opticalspectra are also more common in galaxy pairs, however in constrast to the radiatively-efficient AGN, this enhancement does not depend on separation (Ellison et al. 2015),suggesting that fueling is not directly related to galaxy interactions. To probe the endstages of the merger sequence, one can search for galaxies with double nuclei throughcareful analysis of optical images. These double nuclei show a clear enhancement in thefrequency of X-ray AGN (Lackner et al. 2014), further supporting a connection betweenAGN and mergers.Despite these clear connections between AGN and mergers, a statistical analysis of theAGN population suggests that mergers are associated with only only a minority ( ∼ right ).
4. Obscured AGN and the evolutionary sequence
In studying the population of luminous AGN fueled by major mergers, it is commonto invoke an evolutionary sequence in which the merger produces a powerful, dust ob-scured starburst, followed by a period of powerful obscured AGN activity and finally byan unobscured quasar (e.g., Sanders et al. 1988; Hopkins et al. 2008). In this scenario,obscured quasars represent an important phase in the life of massive galaxies. With sensi-tive mid-IR observations from
Spitzer and
WISE , we have now identified large numbers ofluminous obscured quasars with little or no rest-frame optical emission from the nucleus(e.g., Hickox et al. 2007; Stern et al. 2012; Assef et al. 2013, 2015). X-ray observations(particularly with
NuSTAR ) suggest these obscured quasars are heavily buried or evenCompton-thick (e.g., Stern et al. 2014; Lansbury et al. 2014, 2015), and
Herschel obser-vations suggest that these sources are associated with enhanced star formation (Chenet al. 2015).In addition to these heavily buried quasars, it is particularly interesting to study Hickox et al.quasars that are emerging from the dust. These show intermediate levels of dust extinc-tion, and so are detectable as luminous broad-line AGN but with significant reddening oftheir optical and UV emission. Samples of these ”red QSOs” have been identified usingnear-IR observations in concert with the radio (e.g., Glikman et al. 2007, 2012, 2013),mid-IR (e.g., Banerji et al. 2012, 2013, 2015), and X-rays (e.g., Brusa et al. 2005, 2010).These sources tend to be among the most luminous quasars known (Glikman et al. 2012;Banerji et al. 2015), are characteristically found in ongoing major mergers (e.g., Urrutiaet al. 2008; Glikman et al. 2015, Figure 3), and often show powerful outflows (Brusaet al. 2015a,b; Perna et al. 2015a,b). These characteristics are consistent with their iden-tification as a population in transition between a deeply buried, luminous AGN and anunobscured quasar phase.Because this transitional phase is short-lived and therefore rare, identifying a largesample of reddened QSOs requires the large volumes probed by wide-area surveys. X-rayobservations are particularly powerful for detecting these AGN, and an excellent resourceis the X-ray data set in the SDSS Stripe 82 region (LaMassa et al. 2013, 2015b). The X-ray surveys in Stripe 82 combine archival data with dedicated
XMM-Newton pointings,and contain over 6000 sources over an area of more than 31 deg . The vast majority ofthese sources have multiwavelength counterparts, and ≈
30% have spectroscopic redshifts.Using this large data set, samples of candidate red QSOs have been identified based onoptical, near-IR, and mid-IR properties (e.g., LaMassa et al. 2015a). Near-IR spectro-scopic surveys of these targets are currently underway (LaMassa et al.
5. AGN clustering and dark matter halos
Another valuable technique for studying the evolution of SMBHs is to connect AGNpopulations to their parent dark matter halos via measurements of spatial clustering(see Cappelluti et al. 2012 for a comprehensive review of X-ray AGN clustering). Thegrowth of DM halos is well understood from simulations and analytic theory (e.g., Shethet al. 2001; Tinker et al. 2008), so knowledge of how AGN populate DM halos providesa powerful constraint on models of galaxy formation (e.g., Fanidakis et al. 2013). Mea-surements of the linear clustering amplitude of optical quasars have shown that theyare found in halos of constant mass 10 –10 M (cid:12) at all redshifts (e.g., Croom et al.2005; Myers et al. 2007, Figure 4). Halos of this mass have the highest ratios of stellarmass to dark matter mass (e.g., Moster et al. 2010) and are also the sites of powerful,dust-obscured high-redshift starbursts (submillimeter galaxies; Hickox et al. 2012). Thissuggests that the maximum BH growth occurs in the same systems in which SF is themost efficient. In contrast, slowly-growing, mechanically-dominated AGN are found inhalos of mass > M (cid:12) (e.g., Hickox et al. 2009; Mendez et al. 2015, Figure 4), forwhich virial temperatures of the intergalactic gas are higher, star formation rates arelower, and mechanical energy input is required to offset cooling in the centers of thehalos that would produce further star formation (e.g., Bower et al. 2006; Croton et al.2006).Spatial clustering also provides an independent test of quasar evolutionary models bycomparing the halo masses of obscured and unobscured sources. In the simplest ”unifiedscenario” in which AGN obscuration is due only to orientation (e.g., Netzer 2015), onewould expect no difference in halo mass. However in evolutionary scenarios in whichobscured quasars evolve into unobscured sources, the halo masses of the two types ofquasars can differ. The clustering of obscured quasars identified with Spitzer and
WISEM 6. AGN host galaxies and structures X-rayIR QSOs hot halo formation l og ( H a l o m a ss [ h - M O • ] ) maximal quenchingRadio star-forming systems“exhausted” systems Figure 4.
Illustration of the evolution of dark matter halo mass versus redshift for AGN pop-ulations (Alexander & Hickox 2012). Lines show the evolution of halo mass with redshift forindividual halos. Highlighted is the region of maximum “quenching”, in which halos transitionfrom having large reservoirs of cold gas to being dominated by virialized hot atmospheres. Thegray points show halo masses of optically-bright quasars derived from autocorrelation measure-ments (see Alexander & Hickox 2012 for references). The colored points show the halo massesfor radio, X-ray, and infrared-selected AGN at z ∼ . suggests that they reside in higher mass halos than similar unobscured quasars (e.g.,Hickox et al. 2011; Donoso et al. 2014; DiPompeo et al. 2014, 2015a). However, measure-ments of X-ray selected AGN indicate higher clustering for unobscured sources (Allevatoet al. 2014), and other studies of lower-luminosity IR-selected AGN show no such de-pendence (Mendez et al. 2015). A new, independent tool for studying quasar clusteringcomes from the cross-correlation of quasar positions with lensing maps derived fromthe cosmic microwave background (e.g., Sherwin et al. 2012; Geach et al. 2013); theseanalyses appear to confirm higher host halo masses for obscured quasars (DiPompeoet al. 2015b,a). However, these lensing and clustering results still have relatively largeuncertainties, motivating higher-precision measurements in the future. The relationshipbetween clustering and obscuration thus remains an interesting open question.Recent studies have begun to explore not only the mass of the DM halos that host AGN,but how AGN are distributed within those halos. This work uses the halo occupationdistribution (HOD) formalism, which has proven successful for studying the clusteringof galaxies (e.g., Berlind & Weinberg 2002; Zehavi et al. 2011). Pioneering studies ofthe AGN HOD confirm that AGN are found in halos with average masses 10 –10 M (cid:12) , but suggest that the host halos can span a wide range in mass (e.g., Miyaji et al.2011; Krumpe et al. 2012; Richardson et al. 2013; Shen et al. 2013). HOD analyses ofclustering (Starikova et al. 2011; Richardson et al. 2012, 2013; Shen et al. 2013) anddirect measurements of AGN occupation in groups (Allevato et al. 2012; Silverman et al.2014) also indicate that central galaxies are more likely to host an AGN, although someclustering studies are consistent with a large number of satellites (e.g., Miyaji et al. 2011;Krumpe et al. 2015). There are further some hints that the number of AGN in satellitesrises slowly or even decreases at large halo mass (e.g., Khabiboulline et al. 2014; Krumpe Hickox et al. Figure 5.
Left:
Estimated four-year soft (0.5–2 keV) band sensitivity eRASS in Galactic coor-dinates (Kolodzig et al. 2013b).
Right:
Predicted numbers of AGN detected in the soft band byeRASS as a function of redshift, in five bins of X-ray luminosity, for an area of 14,000 deg , orapproximately the coverage of SDSS (Kolodzig et al. 2013b). et al. 2015), in contrast with the behavior of inactive galaxies, which may provide animportant clue to the process of AGN fueling. While AGN HOD studies are currentlychallenging due to small size of AGN samples and thus limited statistical power, theyrepresent the first step in understanding the complete connection between AGN activityand host DM structures, and provide strong motivation for future generations of largeextragalactic surveys that can identify and characterize large numbers of AGN.
6. The future of large AGN surveys
In the coming years, a number of large-scale surveys over a wide area will dramaticallyexpand the samples of AGN. Wide-area X-ray surveys can efficiently and unambiguouslyidentify AGN with limited contamination from host galaxies, and even relatively shortexposures with current observatories can produce large AGN samples comprising thou-sands of sources, as evidenced by the the wide-area
Chandra
XBo¨otes (Murray et al.2005; Kenter et al. 2005), and
XMM –XXL (Pierre & XXL Consortium 2014), and
XMM
Stripe 82X (LaMassa et al. 2015b) surveys. The next generation wide-area X-ray surveywill be the eRosita All Sky Survey (eRASS), to be carried out by the eROSITA instru-ment on the Spectrum-Roentgen-Gamma spacecraft (Predehl et al. 2007; Merloni et al.2012; Figure 5). eRASS is expected to detect ∼ × AGN over 34,000 deg with amedian redshift z ∼ > AGN at z >
3, increasing the known samples of X-rayAGN by more than an order of magnitude (Kolodzig et al. 2013b). This sample will en-able precision studies of the AGN luminosity function and clustering, using photometricredshifts from wide-field imaging surveys and spectroscopic redshifts from the SPIDERSproject (part of SDSS-III Eisenstein et al. 2011, covering > ∼ >
10) measurements of AGN clustering amplitudein small bins of redshift or luminosity. With sufficiently large numbers of accurate red-shifts, eRASS will also enable the first measurement with X-ray AGN of baryon acousticoscillations (Kolodzig et al. 2013a; H¨utsi et al. 2014).Beyond eROSITA, SDSS-III, and 4MOST, future wide-field survey instruments includethe Athena X-ray mission (Barcons et al. 2015), the Euclid (Laureijs et al. 2012) andWFIRST (Content et al. 2013) optical and near-IR satellite missions, as well as the Sub-aru Prime Focus Spectrograph, which will carry out a wide, deep near-IR spectroscopic
M 6. AGN host galaxies and structures ∼ z ∼ Acknowledgements
R.C.H. acknowledges support from an Alfred P. Sloan Research Fellowship, a Dart-mouth Class of 1962 Faculty Fellowship, the National Science Foundation via grantnumbers 1211096 and 1515364, and NASA through ADAP award NNX12AE38G. Weare grateful to the organizers for an enjoyable and stimulating Focus Meeting and theIAU for organizing an excellent General Assembly.
References
Aird, J., et al. 2012, ApJ, 746, 90Alexander, D. M. & Hickox, R. C. 2012, New Astron. Revs, 56, 93Allevato, V., et al. 2014, ApJ, 796, 4Allevato, V., et al. 2012, ApJ, 758, 47Assef, R. J., et al. 2015, ApJ, 804, 27Assef, R. J., et al. 2013, ApJ, 772, 26Bahcall, J. N., et al. 1997, ApJ, 479, 642Banerji, M., et al. 2015, MNRAS, 447, 3368Banerji, M., et al. 2012, MNRAS, 427, 2275Banerji, M., et al. 2013, MNRAS, 429, L55Barcons, X., et al. 2015, Journal of Physics Conference Series, 610, 012008Berlind, A. A. & Weinberg, D. H. 2002, ApJ, 575, 587Blanton, M. R. 2006, ApJ, 648, 268Booth, C. M. & Schaye, J. 2011, MNRAS, 413, 1158Bower, R. G., et al. 2006, MNRAS, 370, 645Brusa, M., et al. 2015a, MNRAS, 446, 2394Brusa, M., et al. 2010, ApJ, 716, 348Brusa, M., et al. 2005, A&A, 432, 69Brusa, M., et al. 2015b, A&A, 578, A11Cappelluti, N., Allevato, V., & Finoguenov, A. 2012, Advances in Astronomy, 2012Cardamone, C. N., et al. 2010, ApJ, 721, L38Chen, C.-T. J., et al. 2013, ApJ, 773, 3Chen, C.-T. J., et al. 2015, ApJ, 802, 50Cisternas, M., et al. 2011, ApJ, 726, 57Coil, A. L., et al. 2006, ApJ, 644, 671Content, D., et al. 2013, in SPIE Conference Series, Vol. 8860Croom, S. M., et al. 2005, MNRAS, 356, 415Croton, D. J., et al. 2006, MNRAS, 365, 11de Jong, R. S., et al. 2014, in SPIE Conference Series, Vol. 9147Di Matteo, T., Springel, V., & Hernquist, L. 2005, Nature, 433, 604Diamond-Stanic, A. M., et al. 2012, ApJ, 755, L26DiPompeo, M. A., et al. 2015a, MNRASDiPompeo, M. A., et al. 2014, MNRAS, 442, 3443DiPompeo, M. A., et al. 2015b, MNRAS, 446, 3492Donoso, E., et al. 2014, ApJ, 789, 44Eisenstein, D. J., et al. 2011, AJ, 142, 72Ellison, S. L., Patton, D. R., & Hickox, R. C. 2015, MNRAS, 451, L35Ellison, S. L., et al. 2008, AJ, 135, 1877
Fabian, A. C. 2012, ARAA, 50, 455Fanidakis, N., et al. 2013, MNRAS, 435, 679Feruglio, C., et al. 2015, A&A, 583, A99Gabor, J. M. & Bournaud, F. 2013, MNRAS, 434, 606—. 2014, MNRAS, 441, 1615Gabor, J. M., et al. 2009, ApJ, 691, 705Geach, J. E., et al. 2013, ApJ, 776, L41Geach, J. E., et al. 2014, Nature, 516, 68Glikman, E., et al. 2007, ApJ, 667, 673Glikman, E., et al. 2015, ApJ, 806, 218Glikman, E., et al. 2012, ApJ, 757, 51Glikman, E., et al. 2013, ApJ, 778, 127Goulding, A. D., et al. 2014, ApJ, 783, 40Grogin, N. A., et al. 2005, ApJ, 627, L97Hainline, K. N., et al. 2013, ApJ, 774, 145Hainline, K. N., et al. 2014, ApJ, 787, 65Harrison, C. M., et al. 2014, MNRAS, 441, 3306Harrison, F. A., et al. 2013, ApJ, 770, 103Hickox, R. C., et al. 2007, ApJ, 671, 1365Hickox, R. C., et al. 2009, ApJ, 696, 891Hickox, R. C., et al. 2014, ApJ, 782, 9Hickox, R. C., et al. 2011, ApJ, 731, 117Hickox, R. C., et al. 2012, MNRAS, 421, 284Hopkins, P. F., et al. 2008, ApJS, 175, 356H¨utsi, G., et al. 2014, A&A, 572, A28Jansen, F., et al. 2001, A&A, 365, L1Kampczyk, P., et al. 2013, ApJ, 762, 43Kartaltepe, J. S., et al. 2012, ApJ, 757, 23Keel, W. C., et al. 2015, AJ, 149, 155Kenter, A., et al. 2005, ApJS, 161, 9Khabiboulline, E. T., et al. 2014, ApJ, 795, 62Kocevski, D. D., et al. 2012, ApJ, 744, 148Kolodzig, A., et al. 2013a, A&A, 558, A90Kolodzig, A., et al. 2013b, A&A, 558, A89Kormendy, J. & Ho, L. C. 2013, ARAA, 51, 511Koss, M., et al. 2010, ApJ, 716, L125Krumpe, M., et al. 2012, ApJ, 746, 1Krumpe, M., et al. 2015, ApJ, 815, 21Lackner, C. N., et al. 2014, AJ, 148, 137LaMassa, S. M., et al. 2015a, ApJ submitted (arXiv:1511.02883)LaMassa, S. M., et al. 2015b, ApJ submitted (arXiv:1510.00852)LaMassa, S. M., et al. 2013, MNRAS, 436, 3581Lansbury, G. B., et al. 2014, ApJ, 785, 17Lansbury, G. B., et al. 2015, ApJ, 809, 115Laureijs, R., et al. 2012, in SPIE Conference Series, Vol. 8442McConnell, N. J. & Ma, C.-P. 2013, ApJ, 764, 184Mechtley, M., et al. 2015, ArXiv e-printsMendez, A. J., et al. 2015, ApJ submitted (arXiv:1504.06284)Merloni, A. & Heinz, et al. 2013, Evolution of Active Galactic Nuclei, ed. T. D. Oswalt & W. C.Keel, 503Merloni, A., et al. 2012, eROSITA Science Book: Mapping the Structure of the Energetic Uni-verse (arXiv:1209.3114)Miyaji, T., et al. 2011, ApJ, 726, 83Moster, B. P., et al. 2010, ApJ, 710, 903
M 6. AGN host galaxies and structures11