A local baseline of the black hole mass scaling relations for active galaxies. IV. Correlations between M_{\rm BH} and host galaxy σ, stellar mass, and luminosity
Vardha N. Bennert, Tommaso Treu, Xuheng Ding, Isak Stomberg, Simon Birrer, Tomas Snyder, Matthew A. Malkan, Andrew W. Stephens, Matthew W. Auger
DDraft version January 27, 2021
Typeset using L A TEX twocolumn style in AASTeX63
A local baseline of the black hole mass scaling relations for active galaxies. IV. Correlations between M BH and host galaxy σ , stellar mass, and luminosity. Vardha N. Bennert, Tommaso Treu, Xuheng Ding, Isak Stomberg,
1, 3, 4, 5
Simon Birrer, Tomas Snyder, Matthew A. Malkan, Andrew W. Stephens, and Matthew W. Auger Physics Department, California Polytechnic State University, San Luis Obispo, CA 93407, USA Department of Physics, University of California, Los Angeles, CA 90095, USA Department of Physics, KTH, Royal Institute of Technology, Stockholm, Sweden Universit¨at Hamburg, Department of Physics, 20355 Hamburg, Germany Deutsches Elektronen-Synchrotron DESY, 22607 Hamburg, Germany Kavli Institute for Particle Astrophysics and Cosmology and Department of Physics, Stanford University, Stanford, CA 94305, USA Gemini Observatory/NSF’s NOIRLab, 670 N. A’ohoku Place, Hilo, Hawai’i, 96720, USA Institute of Astronomy, Madingley Road, Cambridge CB3 0HA, UK (Received; Revised; Accepted)
ABSTRACTThe tight correlations between the mass of supermassive black holes ( M BH ) and their host-galaxyproperties have been of great interest to the astrophysical community, but a clear understanding of theirorigin and fundamental drivers still eludes us. The local relations for active galaxies are interestingin their own right and form the foundation for any evolutionary study over cosmic time. We presentHubble Space Telescope optical imaging of a sample of 66 local active galactic nuclei (AGNs); for 14objects, we also obtained Gemini near-infrared images. We use state of the art methods to performsurface photometry of the AGN host galaxies, decomposing them in spheroid, disk and bar (whenpresent) and inferring the luminosity and stellar mass of the components. We combine this informationwith spatially-resolved kinematics obtained at the Keck Telescopes to study the correlations between M BH (determined from single-epoch virial estimators) and host galaxy properties. Our sample extendsthe correlation found for quiescent galaxies down to M BH ∼ M (cid:12) , along a consistent line. Thecorrelations are uniformly tight for our AGN sample, with intrinsic scatter 0 . − . -10 M (cid:12) regime, with data of sufficient quality. The M BH - σ relation is also inexcellent agreement with that of AGN with M BH obtained from reverberation mapping, providing anindirect validation of single-epoch virial estimators of M BH . Keywords: accretion, accretion disks − black hole physics − galaxies: active − galaxies: evolution − galaxies: Seyfert INTRODUCTIONWhen growing through accretion, supermassive blackholes (BHs) can be seen as bright nuclei in active galax-ies (AGNs). The observed relations between the massof the BH ( M BH ) and the properties of the host-galaxyspheroid such as luminosity, stellar mass and stellar- Corresponding author: Vardha N. [email protected] velocity dispersion σ , are thought to result from the co-evolution between BHs and galaxies (for a review see,e.g., Kormendy & Ho 2013; Graham 2016). Such a co-evolution is either regulated by AGN feedback (e.g., DiMatteo et al. 2005; Croton 2006; Dubois et al. 2013,2016; DeGraf et al. 2015; Hopkins et al. 2016), or hi-erarchical assembly of M BH and stellar mass throughgalaxy merging (e.g., Peng 2007; Hirschmann et al. 2010;Jahnke & Maccio 2011). To shed light on the origin ofthese relations, recent years have seen an explosion ofobservational studies both in the local Universe (e.g., a r X i v : . [ a s t r o - ph . GA ] J a n Bennert et al.
Ferrarese & Ford 2005; Greene & Ho 2006; G¨ultekin etal. 2009; Bennert et al. 2011a; Kormendy et al. 2011;Beifiori et al. 2012; L¨asker et al. 2016; Davis et al. 2018;Sahu et al. 2019) and as a function of cosmic history (e.g.Treu et al. 2004; Peng et al. 2006a,b; Woo et al. 2006;Salviander et al. 2007; Riechers et al. 2009; Jahnke et al.2009; Bennert et al. 2010; Decarli et al. 2010; Merloni etal. 2010; Bennert et al. 2011b; Park et al. 2015; Sextonet al. 2019; Silverman et al. 2019; Ding et al. 2020).By necessity, all studies beyond the local Universe fo-cus on broad-line (or type-1) AGNs (BLAGNs). ForBLAGNs, M BH can be estimated to within a factorof 2-3 using empirically calibrated relations based ona sample of reverberation-mapped AGNs. Reverbera-tion mapping (RM) is a technique that studies the timedelay between the variability of the accretion disk andthe response of ionized gas in the vicinity of the BH, thebroad-line region (BLR) (e.g., Wandel et al. 1999; Woo& Urry 2002; Vestergaard 2002; Vestergaard & Peterson2006; McGill et al. 2008). Using light-travel time argu-ments, the time delay translates into a size of the BLR.Combining the size with the Doppler-broadened widthof the emission lines (e.g., the Hydrogen Balmer seriesin the optical) results in an estimate of the M BH up toan unknown factor that depends on the geometry andkinematics of the gas clouds. Traditionally, this factor f (also known as virial factor) has been derived as a sam-ple average by matching the scaling relation between M BH and (spheroid) stellar-velocity dispersion σ of theRM AGNs with that of quiescent galaxies (e.g., Onken etal. 2004; Park et al. 2012; Woo et al. 2010, 2015). Morerecently, dynamical modeling of RM data has been usedto constrain both geometry and kinematics of the BLRand thus determine M BH for individual objects, findingconsistent results (e.g., Brewer et al. 2011; Pancoast etal. 2011; Li et al. 2013; Pancoast et al. 2018; Williams etal. 2018, 2020). While RM is time-consuming, the RMAGN sample revealed a relation between BLR size andAGN luminosity that can be used to estimate M BH forBLAGNs from one spectrum, known as the single-epochmethod. In the single-epoch virial estimation, the widthof broad emission lines is combined with the AGN lumi-nosity which serves as a proxy for BLR size. As such,the RM AGN sample serves as a M BH calibrator beyondthe local Universe. The single-epoch method has beenused for virial mass estimates of hundreds of thousandsof AGNs (e.g., Rakshit et al. 2020), across cosmic his-tory (e.g., Mortlock et al. 2011), to study the cosmicevolution of the M BH scaling relations (e.g., Treu et al.2004; Peng et al. 2006a; Woo et al. 2006; Bennert et al.2010; Merloni et al. 2010; Park et al. 2015; Ding et al. 2020) and distribution of Eddington ratios (e.g., Shen2013).Studies of the evolution of the M BH -host-galaxy scal-ing relations with redshift constrain theoretical inter-pretations and shed light onto their origin (e.g., Croton2006; Hopkins et al. 2007); however, they depend on ourunderstanding of the slope and intrinsic scatter of localrelations, in particular those for active galaxies. More-over, studying dependencies of the correlations on bulgestructure and other morphological components at high-redshifts is difficult if not impossible, especially giventhe presence of the bright AGN point source in the cen-ter. Late-type galaxies are often known to host pseudo-bulges, characterized by exponential light profiles, on-going star formation or starbursts, and nuclear bars. Itis generally believed that they have evolved secularlythrough dissipative processes rather than mergers (e.g.,Courteau et al. 1996; Kormendy & Kennicutt 2004).Classical bulges, in contrast, are thought of as centrallyconcentrated, mostly red and quiescent, merger-inducedsystems. Pseudo-bulges and minor mergers provide avaluable test of some hypotheses for the origin of the M BH scaling relations: If they lie off the relations, ma-jor mergers can be considered the fundamental driver(Peng 2007; Hirschmann et al. 2010; Jahnke & Maccio2011; Kormendy & Ho 2013); if they lie on the relation,as found by our results based on SDSS images (Bennertet al. 2015), it could indicate that secular evolution hasa synchronizing effect, growing BHs and bulges simulta-neously at a small but steady rate for late-type galaxies(Cisternas et al. 2011a,b).This paper is the last of a series aimed at creatinga robust local baseline of the M BH scaling relations ofBLAGNs for comparison with high redshift studies. Weselected a sample of ∼
100 Seyfert-1 galaxies from SDSS(0.02 ≤ z ≤ M BH > M (cid:12) ) based on their broadH β emission in the same fashion as high-redshift sam-ples used for evolutionary studies (Bennert et al. 2010;Park et al. 2015; Ding et al. 2020), allowing for a di-rect comparison. The majority of AGNs ( ∼ M BH measure-ment), and spheroid stellar masses. In paper II (Har-ris et al. 2012), high-quality long-slit Keck/LRIS spec-tra provided both M BH estimates as well as accurate lack hole mass scaling relations for local active galaxies σ ) and ro-tation curves.Given the wide range of M BH , host-galaxy morpholo-gies, and stellar masses, our sample is well suited todetermine the slope and intrinsic scatter of the localscaling relations and to study dependencies on otherparameters such as bulge structure and mergers. How-ever, relying on low-quality optical photometry (such asSDSS) whose insufficient angular resolution and limitedsensitivity to dust extinction significantly increase obser-vational scatter in the M BH scaling relations, ultimatelylimits conclusive results. High-resolution images are es-sential to resolve (pseudo-) bulges, given that roughlyhalf of all objects have bulge effective radii smaller than ∼ (cid:48)(cid:48) (corresponding to ∼ M BH scaling relations.In this paper, to overcome these problems and to ob-tain high-quality host-galaxy images, we took a two-pronged approach. A sub-set of the parent sample (15objects), selected to cover a wide range of morphologies(as based on SDSS images), was observed with the NearInfraRed Imager and spectrograph (NIRI) on GeminiNorth. Gemini-NIRI was chosen (i) for its high-spatialresolution (instrument plus site seeing) to distinguishbetween classical and pseudo-bulges in the presence ofan AGN point source; (ii) for its large field-of-view(2 (cid:48) × (cid:48) at f/6) to measure the surface brightness profileof these nearby galaxies out to large radii; and (iii) be-cause near-infrared observations maximize the contrastbetween AGN and host and minimize dust extinction,revealing the presence of (pseudo-) bulges, bars and (mi-nor) mergers. The reduced dust extinction also makesNIR luminosities a better tracer of stellar mass. At thesame time, the parent sample was part of a Hubble SpaceTelescope (HST) snapshot (SNAP) program (PI Ben-nert). HST images provide both a high spatial resolutionand a stable point-spread function (PSF). WFC3/UVISwas used with broad-band filter F814W to maximizethe contrast between AGN and host, avoiding strongAGN emission lines, while taking full advantage of thehigh resolution of UVIS. Compared to existing SDSSimages, the HST images have a factor of ∼
40 increase in resolution. A total of 66 objects were observed withHST, 14 of which also have Gemini images. Gemini andHST images naturally complement each other and to-gether provide a long wavelength range for stellar-massdetermination for overlapping objects. By construction,SDSS images in five filters are available for all objects tofurther assist in constraining stellar masses. Combiningthe high-quality spectroscopic data (paper II) with high-quality imaging provides a representative spectral andspatial coverage of supermassive BHs and their hosts fora detailed mapping of the local M BH scaling relations foractive galaxies and their underlying drivers.The paper is organized as follows. Section 2 sum-marizes the sample selection, HST and Gemini obser-vations, and data reduction. Section 3 describes theanalysis and derived quantities. Section 4 presents host-galaxy morphologies and discusses the resulting M BH -scaling relations. Section 5 concludes with a summary.Throughout this paper, magnitudes are given in ABmagnitudes. For conversion to luminosities, absolutesolar magnitudes were taken from Willmer (2018) and aHubble constant of H = 70 km s − and a flat Universewith a cosmological constant of Ω λ = 0.7 are assumed. SAMPLE SELECTION, OBSERVATIONS, ANDDATA REDUCTION2.1.
Sample selection
The parent sample is 102 type-1 Seyfert galaxies se-lected from the Sloan Digital Sky Survey (SDSS) datarelease six (Adelman-McCarthy et al. 2008) based onredshift (0.02 ≤ z ≤ M BH ( > M (cid:12) ), and ob-served with Keck/LRIS, presented in detail in papers I,II and III in this series (Bennert et al. 2011a; Harris et al.2012; Bennert et al. 2015). Accurate spatially-resolvedstellar-velocity dispersions were obtained for 84 objects(paper II) and formed the sample for our HST snap-shot (SNAP) program. 15 objects with a wide varietyof host-galaxy properties, as determined from SDSS im-ages (Bennert et al. 2015), were observed with NIRI onGemini North (PI: Bennert; program ID GN-2016B-Q-33). 68 objects were observed as part of HST SNAP,although we were unable to determine a robust M BH for2 of them due to a lack of broad H β in the Keck spec-trum (despite it being present in prior SDSS spectra;Runco et al. 2016). Of the remaining 66, 14 overlapwith the Gemini sample and so we include here onlythose 14 objects observed with both HST and Gemini.Fully-reduced SDSS images are available for all objectsthrough the SDSS archive. Sample properties (coordi-nates and redshift) and derived quantities can be foundin Table 1. Bennert et al. T a b l e . S a m p l e a ndd e r i v e d q u a n t i t i e s O b j ec t R . A . D ec l. z P A F r a m e s i ze M B H σ (cid:63) M s ph , d y n L I , s ph L I , s ph + b a r M s ph , (cid:63) M s ph + b a r , (cid:63) H o s t Sph e r o i d ( J )( J )( E o f N )( ” )( l og M (cid:12) )( k m s − )( l og M (cid:12) )( l og L (cid:12) )( l og L (cid:12) )( l og M (cid:12) )( l og M (cid:12) ) ( )( )( )( )( )( )( )( )( )( )( )( )( )( )( ) - . - . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P - . - . . . . . . . . ...... B D B P + . + . . . . . . . . . ± . . ± . B D B P - . - . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P - . - . . . . . . . ... . ± . ... B D C - . - . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . ...... B D B P + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B C + . + . . . . . . . . . ± . . ± . B D B P T a b l e c o n t i n u e d lack hole mass scaling relations for local active galaxies T a b l e ( c o n t i n u e d ) O b j ec t R . A . D ec l. z P A F r a m e s i ze M B H σ (cid:63) M s ph , d y n L I , s ph L I , s ph + b a r M s ph , (cid:63) M s ph + b a r , (cid:63) H o s t Sph e r o i d ( J )( J )( E o f N )( ” )( l og M (cid:12) )( k m s − )( l og M (cid:12) )( l og L (cid:12) )( l og L (cid:12) )( l og M (cid:12) )( l og M (cid:12) ) ( )( )( )( )( )( )( )( )( )( )( )( )( )( )( ) + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D P + . + . . . . . . . . ...... B D B P + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . . . ± . . ± . B D B P - . - . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D P + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . . . ± . . ± . B D B C + . + . . . . . . . ... . ± . ... B D P + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C - . - . . . . . . . ... . ± . ... B D C - . - . . . . . . . ... . ± . ... B D C - . - . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . . . ± . . ± . B D B P + . + . . . . . . . ... . ± . ... B D C T a b l e c o n t i n u e d Bennert et al. T a b l e ( c o n t i n u e d ) O b j ec t R . A . D ec l. z P A F r a m e s i ze M B H σ (cid:63) M s ph , d y n L I , s ph L I , s ph + b a r M s ph , (cid:63) M s ph + b a r , (cid:63) H o s t Sph e r o i d ( J )( J )( E o f N )( ” )( l og M (cid:12) )( k m s − )( l og M (cid:12) )( l og L (cid:12) )( l og L (cid:12) )( l og M (cid:12) )( l og M (cid:12) ) ( )( )( )( )( )( )( )( )( )( )( )( )( )( )( ) + . + . . . . . . . ... . ± . ... B D C + . + . . . . . . . ... . ± . ... B D C N o t e — C o l. ( ) : T a r g e t I D u s e d t h r o u g h o u tt h e t e x t( b a s e d o n R . A . a ndd ec li n a t i o n ) . C o l. ( ) : R i g h t a s ce n s i o n i nh o u r s , m i nu t e s a nd s ec o nd s . C o l. ( ) : D ec li n a t i o n i nd e g r ee s , a r c m i nu t e s a nd a r c s ec o nd s . C o l. ( ) : R e d s h i f t f r o m S D SSd a t a r e l e a s e s e v e n ( A b a z a j i a n e t a l. ) . C o l. ( ) : P o s i t i o n a n g l e ( P A ) s h o w n i n F i g s . - ( i nd e g r ee s E o f N ) . N o t e t h a tt h i s i s t h e P A o f t h e i m ag e , n o tt h e h o s t ga l a xy . C o l. ( ) : F r a m e s i ze o f i m ag e s h o w n i n F i g s . - ( i n a r c s ec o nd ,i n x a nd y ) . C o l. ( ) : L oga r i t h m o f M B H ( s o l a r un i t s ) ( un ce r t a i n t y o f . d e x ) . C o l. ( ) : S t e ll a r - v e l o c i t y d i s p e r s i o n w i t h i n s ph e r o i d e ff ec t i v e r a d i u s ( i n k m s − )( un ce r t a i n t y o f . d e x ) . C o l. ( ) : L oga r i t h m o f s ph e r o i dd y n a m i c a l m a ss ( s o l a r un i t s )( un ce r t a i n t y o f . d e x ) . C o l. ( ) : L oga r i t h m o f s ph e r o i d I - b a nd l u m i n o s i t y ( s o l a r un i t s )( un ce r t a i n t y o f . d e x ) . C o l. ( ) : L oga r i t h m o f s ph e r o i d + b a r I - b a nd l u m i n o s i t y ( s o l a r un i t s )( un ce r t a i n t y o f . d e x ) . C o l. ( ) : L oga r i t h m o f s ph e r o i d s t e ll a r m a ss ( s o l a r un i t s ) . C o l. ( ) : L oga r i t h m o f s ph e r o i d + b a r s t e ll a r m a ss ( s o l a r un i t s ) . C o l. ( ) : H o s t - ga l a xy fi t( B : s ph e r o i d o n l y , B D : s ph e r o i d + d i s k , B D B : s ph e r o i d + d i s k + b a r ) . C o l. ( ) : Sph e r o i d c o m p o n e n t : C = c l a ss i c a l bu l g e ; P = p s e ud o - bu l g e . lack hole mass scaling relations for local active galaxies HST observations and data reduction
84 objects were part of an HST SNAP program (HSTGO 15215 and HST GO 16014, PI Bennert; Cycles 25-27), yielding images for 68 AGN host galaxies, a com-pletion rate of almost 80% (significantly higher thanthe 30% typical for SNAP programs). To obtain thedynamic range needed for an accurate decompositionof the host galaxy and the AGN, the long exposures(400 seconds) were complemented by short, unsaturatedones (between 20-100 seconds, depending on the object’sbrightness). To avoid buffer dump (which occurred dur-ing the long exposures), a sequence of one short andone long exposure at the same location was followed byanother sequence of one long and one short exposureat a dithered location. POS TARG was used to set upa dither pattern manually that corresponds to the de-fault WFC3-UVIS-GAP-LINE (with center UVIS). Full-frame images were read to trace the host-galaxy disksout to the background and to obtain field stars for PSFfitting of the strong AGN point source in the center.Data processed through the standard WFC3 cali-bration pipeline were retrieved from the HST archive.L.A. Cosmic (Laplacian Cosmic Ray Identification) (vanDokkum 2001) was run to remove cosmic rays. Alllong exposures were carefully checked for saturation,especially of the bright AGN point source. For ob-jects with saturated pixels, the short exposures takenat the same dither location as the long ones werescaled according to exposure time and used to replacethe saturated pixels. Pyraf package astrodrizzle wasthen used to combine the two long exposures. A widerange of combinations of the final drizzle parametersscale and pixfrac were applied. After careful exami-nation of the images and based on resolution, imagequality and FWHM of the PSF, the following param-eters were adopted: driz sep bits=336, final bits=336,final wcs=yes, final pixfrac=0.9, final scale=0.035, re-sulting in a final pixel scale of 0.035 (cid:48)(cid:48) /pix. For objectswithout any saturated pixels, the long exposures werecombined with astrodrizzle directly in the same way. Fi-nal images are shown in Figures 1-2.2.3.
Gemini observations and data reduction
15 Seyfert-1 galaxies were selected from the parentsample covering a wide range of morphologies (based onSDSS images). All galaxies were observed with NIRI onGemini North with the largest field-of-view (FOV; 2 (cid:48) × (cid:48) at f/6; pixel scale of 0.117 (cid:48)(cid:48) ) in Ks band. Observationshave an average FWHM of 0.33 (cid:48)(cid:48) (0.24 - 0.44 (cid:48)(cid:48) ) and wereobtained at airmass less than 1.5. This image quality is3-4 times better than the SDSS images which have atypical seeing of 1.5 (cid:48)(cid:48) . Exposure times range between 2- 10 seconds per image with 3-6 co-adds and 18-24 imageson source, resulting in a total exposure time of 144-540seconds, depending on the brightness of the object. Thesky positions were observed with guiding enabled andwere carefully selected to include a nearby bright fieldstar. This ensured that we could generate accurate PSFmodels for every galaxy. Details of the observations aregiven in Table 2.Data reduction was performed following standard pro-cedures using the Gemini IRAF package customized forNIRI and included dark subtraction and flat fielding us-ing off-target exposures. Flux calibration was obtainedby standard IRAF photometry of UKIRT faint standardstars observed directly before or after the science im-ages. Absolute magnitudes take into account extinction(Schlafly & Finkbeiner 2011); Ks-band luminosities weredetermined assuming an absolute Ks-band magnitude ofthe Sun of M Ks = 5.08 (Willmer 2018) Resulting imagesare shown in Figure 3. DERIVED QUANTITIES3.1.
Surface photometry
To perform a detailed 2D host-galaxy fitting, we usethe public image analysis software lenstronomy (Birreret al. 2018). Lenstronomy supersedes GALFIT (Peng etal. 2002) by applying an MCMC technique to providerealistic errors and explore the covariance between thevarious model parameters. It allows for a more generalsurface brightness reconstruction possible with a largenon-parametric basis set; the coefficients are determinedthrough a linear minimization rather than a non-linearparameter fitting (Birrer et al. 2015). Also, iterativePSF reconstruction is possible and allows one to incor-porate residual uncertainties due to PSF mismatch intothe analysis. While lenstronomy was originally devel-oped for galaxy-scale strong gravitational lensing, it hasa much broader application, including general 2D galaxydecompositions.A well subtracted background and a matching PSF isimportant for obtaining reliable host galaxy propertiesfrom 2D surface photometry. We estimate and removethe local background light in 2D space based on theSExtractorBackground algorithm built in the photutilspackage (Python based), which effectively accounts forgradients in the background light distribution. For allobjects, PSF stars were created from suitable stars in thefield-of-view (FOV) of each object, following the criteria:bright, unsaturated star without any nearby objects, lo-cated close to the AGN/center of the FOV and an over- https://github.com/sibirrer/lenstronomy Bennert et al.
Figure 1.
HST UVIS/F814W images of our sample. Position angles and image sizes are listed in table 1. Thanks to the highspatial resolution and S/N data, the wide variety of host-galaxy morphologies can clearly be seen. lack hole mass scaling relations for local active galaxies Figure 2.
Continuation of Figure 1. Bennert et al.
Figure 3.
Gemini NIRI Ks images for 14 galaxies. North is up, East is to the left. Image sizes are listed in table 2. all profile as expected for a PSF star. These individualPSF stars were then recentered and stacked, resultingin a PSF with high signal-to-noise ratio (S/N), centeredin the image. Each individual PSF star was recenteredusing AstroObjectAnalyzer through an iterative inter-polation algorithm. Finally, masks were created to maskany nearby sources. Three objects have close-by neigh-boring galaxies which were fitted simultaneously.The central AGN was fitted by a PSF, the host galax-ies with three different models: (1) a spheroid-onlycomponent (free S´ersic index n ; S´ersic 1963); (2) aspheroid plus disk component (S´ersic index = 1) (3)a spheroid plus disk plus bar component (S´ersic index= 0.5). Based on pre-defined starting parameters andconstraints, lenstronomy determines the maximum like-lihood fit adopting a Particle Swarm Optimizer (PSO)(Kennedy & Eberhart 2001). We use the following lim-its: effective radius r eff for all components: 3 × pixelsize ≤ r eff ≤ × pixel size; spheroid S´ersic index n :1 ≤ n ≤
5. Also, for the spheroid-disk fit, we forcethe disk to be larger than the spheroid and more ellipti- https://github.com/sibirrer/AstroObjectAnalyser cal. Likewise, for the spheroid-disk-bar fit, we force thedisk to be larger than the bar and the spheroid, and thespheroid component to be the most round one of threecomponents.After running various trials, the two main challengesthat we encountered were (i) determining the best val-ues for the PSO chains, to make sure the code convergedand (ii) making sure that the code converged to the trueglobal minimum and not to a local one. The latter maydepend on the starting parameters used, especially ifmore than one component is fitted to the host galaxy,due to degeneracies involved and the high-dimensionalparameter volume. After some experimenting, the fol-lowing procedure was shown to be successful. We chosea PSO chain that guaranteed convergence even for thelargest image size. For a spheroid-only fit, we chose aPSO with 200 particles and 70 iterations, for spheroid-disk and spheroid-disk-bar, a PSO with 300 particlesand 100 iterations. For all fits, as a diagnostic, we dis-played the log (likelihood) of the fit, particle positionand parameter velocity for the different parameters asa function of iteration to ensure that the chain indeedconverged. lack hole mass scaling relations for local active galaxies Table 2.
Gemini Observations.
Object Date of Obs. Exp. time FWHM Frame size(UT) (s) (”) (”)(1) (2) (3) (4) (5)0013 − − − − − Note — Col. (1): Target ID used throughout the text (based onR.A. and declination). Col. (2): Date of observation (UT). Col.(3): Total exposure time in seconds. Col. (4): Full-Width at HalfMaximum (FWHM) of PSF star in arcsecond. Col. (5): Framesize of image shown in Fig. 3 (in arcsecond, in x and y).
First, we ran the spheroid-only fit which was shown tobe robust, i.e. giving the same fitting result, regardlessof starting parameters used. We then used the resultsfrom the spheroid-only fit as starting parameters for thedisk in the subsequent spheroid-disk and spheroid-disk-bar fit. For objects for which there is at least an indi-cation of a visual bar, the bar parameters size, positionangle and ellipticity were carefully determined manu-ally and used as starting parameters for the bar in thespheroid-disk-bar fit. For both the spheroid-disk andspheroid-disk-bar fit, we chose three different startingparameters for the size of the spheroid and the disk, tocover a broader range. For the spheroid effective ra-dius, we chose three different starting parameters: (i)pixel size * 4, (ii) pixel size * 8, and (iii) pixel size *30.For the disk effective radius we chose these three dif-ferent starting parameters: (i) spheroid effective radiusfrom spheroid-only fit, (ii) twice the size used in (i),and (iii) half the size used in (i). When combined, thisyields nine different starting parameters for both thespheroid-disk and spheroid-disk-bar fits. The nine fitswere compared in terms of image residuals and result-ing chi-squared values. In most cases, the nine differentfits yielded identical results, showing the convergenceto a true global minimum. Occasionally, outliers wereidentified through higher chi-square values and/or fromthe residual image and excluded. By careful inspection of the images and final fitting results, we determinedthe best model and fit for all objects. A disk and barwere included in the host-galaxy fitting only if they wereclearly visible in the image and/or if their inclusion sig-nificantly improved the fit (as evidenced by chi squareand residuals), beyond the typical scatter of values seenfor the nine different fits for a given model. We conser-vatively adopt 0.04 dex as uncertainty on the derivedluminosities. Figure 4 shows example fits by lenstron-omy with our chosen procedure. Table 4 in the appendixgives the results of the fitting.Galactic foreground extinction was subtracted basedon dust reddening measurements from Schlafly &Finkbeiner (2011), assuming F814W = 0.61 A V . Mag-nitudes were converted to luminosities, applying a k -correction using Astrolib PySynphot and a Kinney etal. (1996) Sa galaxy template. Note that given the lowredshift of our galaxies, using a different template doesnot significantly change our results. Also, pysynphotgave (V-I) colors less than 1.2 magnitude for all galaxytemplates (elliptical, S0, Sa, or Sb) and thus, the F814Wfilter magnitudes can be considered identical to I-bandmagnitudes (Harris 2018). https://pysynphot.readthedocs.io/en/latest/ Bennert et al.
To derive colors, we fitted the Gemini and SDSS im-ages (in the filters g (cid:48) , r (cid:48) , i (cid:48) and z (cid:48) ) in a similar wayusing lenstronomy. We first used the Gemini images toindependently determine the host galaxy morphology ofeach galaxy, based on both visual inspection of the im-ages and lenstronomy fitting results. They agree wellwith the conclusions reached from the HST fitting. Wethen adopted the same host-galaxy parameters derivedfrom the fitting of the HST images and used lenstron-omy to fit the Gemini and SDSS images, leaving onlythe magnitudes of PSF and host-galaxy components asfree parameters. This gives us magnitudes in 5 or 6different filters for the different host-galaxy componentsspheroid, (pseudo-) bulge, and disk, if present. Dustextinction and k correction were applied.As many literature studies rely on GALFIT, we alsoran GALFIT on the same background-subtracted HSTimages for comparison, using the same PSFs and overallprocedure as for lenstronomy. Overall, the results agree,especially (and not surprisingly) for a single-componentfit (spheroid only), as well as the total host-galaxy mag-nitude (similar conclusions were also reached by Yanget al. 2020). For more complicated models, the effectiveradii for individual objects scatter, but the biggest dif-ference is seen in the spheroid S´ersic index (since diskand bar have fixed S´ersic indices). While the mean ofthe ratio between S´ersic index n as determined fromlenstronomy and GALFIT is still around 1, it scattersgreatly (0.97 ± ± n aloneas an indication of the nature of the spheroid (classicalvs. pseudo-bulge) and in this paper, we use a conserva-tive approach (see discussion in Section 4.1). However,we want to stress that GALFIT tends to need more userinteraction and visual inspections of results to ensure atrue global minimum was reached which we did not dohere. Lenstronomy’s design of semi-linear inversion andPSO resulted in a significant improvement in automa-tion and reduction of labor-intensive work in the fittingprocess relative to GALFIT.3.2. Stellar-velocity dispersion and black hole mass
In the literature, the observed correlation between M BH and σ is generally considered the tightest and thusmost fundamental of the M BH -host-galaxy scaling rela-tions (Tremaine et al. 2002; Beifiori et al. 2012; Sagliaet al. 2016; Shankar et al. 2016; van den Bosch 2016; deNicola et al. 2019). Moreover, it is used to calibrate M BH by matching the M BH - σ relation of RM AGNsto that of quiescent galaxies. Thus, robust measure-ments of σ are essential. There are several definitionsof σ used in the literature, resulting in widely differ- ing measurements depending on aperture size used andhost-galaxy morphology (see paper III of this series Ben-nert et al. 2015), with the most robust being spatially-resolved stellar-velocity dispersions within the effectivespheroid radius. Spatially-resolved stellar-velocity dis-persions were presented in paper II (Harris et al. 2012)based on our Keck spectra. In paper III (Bennert et al.2015), we determined σ from spatially-resolved σ mea-surements integrated within the effective spheroid ra-dius (see equation (1) in paper III; Bennert et al. 2015).However, the effective spheroid radius in paper III wasbased on surface-photometry of SDSS images. We hererepeat the same calculation, now using robust effectivespheroid radii from the HST surface photometry. Whencompared, on average, the σ values are similar (the onesbased on HST radii are larger by 1%), but with a largescatter of 10%, and a couple of individual objects havingchanged by as much as 50%. For 16 objects, the lack ofsufficient spatially-resolved σ measurements hindered arobust determination of σ within the effective spheroidradius and they were excluded here. Thus, the M BH - σ relation presented in section 4.2 includes 50 objects. M BH was determined for the entire sample in Bennertet al. (2015), based on the second moment of the broadH β emission line determined from Keck spectroscopy.The 5100˚A AGN luminosity was used as a proxy forBLR size and combined with the width of H β to esti-mate M BH as in equation (2) in Bennert et al. (2015).In Bennert et al. (2015), a virial factor of log f = 0 . f independentlyby matching the M BH - σ relation to that of Kormendy &Ho (2013). To do so, we first fix the slope of the M BH - σ relation to the one from Kormendy & Ho (2013) andthen adjust log f to match the intercept, resulting inlog f = 0 .
97. A wide spread in virial factors (log f rang-ing between 0.5 and 1.2) has also been found in previousstudies, depending on the choice of different quiescentsamples, fitting methods and M BH range (e.g., Park etal. 2012; Shankar et al. 2019), possibly due to selectioneffects in the local sample of quiescent black holes.3.3. Stellar and dynamical masses
From our surface photometry (Section 3.1), wehave magnitudes for five to six different bands (HSTUVIS/F814W, SDSS g (cid:48) , r (cid:48) , i (cid:48) , z (cid:48) for all objects plusGemini NIRI/Ks for 14 objects) for the different host-galaxy components, (pseudo-) bulge, disk, and bar, ifpresent. To estimate stellar masses from colors, we usea Bayesian stellar-mass estimation code with priors on lack hole mass scaling relations for local active galaxies + data model data - point source normalized residual pixel ( m a g , p i x e l ) datamodelQSOhost0.1 0.2 0.5 1.0 2.0 5.0 arcsec + data model data - point source normalized residual pixel ( m a g , p i x e l ) datamodelQSOhost as 2 components0.1 0.2 0.5 1.0 2.0 5.0 10.0 arcsec + data model data - point source normalized residual pixel ( m a g , p i x e l ) datamodelQSOhost as 2 components0.1 0.2 0.5 1.0 2.0 5.0 10.0 arcsec + data model data - point source normalized residual pixel ( m a g , p i x e l ) datamodelQSOhost as 3 components0.1 0.2 0.5 1.0 2.0 5.0 10.0 arcsec + data model data - point source normalized residual pixel ( m a g , p i x e l ) datamodelQSOhost as 3 components0.1 0.2 0.5 1.0 2.0 5.0 10.0 arcsec Figure 4.
Example fits from lenstronomy. From left to right: HST image (“data”), best fit model derived from lenstronomy(“model”), PSF-subtracted image (“data - Point Source”), residual image after subtraction of best fit model from data (“nor-malized residual”) and surface-brightness profile showing the data as black circles, the model as a blue line, the PSF fittingthe central AGN as an orange line and the host-galaxy total fit as a green line. From top to bottom, five different objects areshown, representing a variety of host-galaxy morphologies: 1043+1105 is an elliptical galaxy fit with a single S´ersic; 1416+0137is a disk galaxy with classical bulge; 1419+0754 is a disk galaxy with pseudo-bulge; 0845+3409 is a barred disk galaxy withclassical bulge; and 1206+4244 is a barred disk galaxy with pseudo-bulge. Bennert et al. age, metallicity and dust content of the galaxy and er-ror bars on the different magnitudes (Auger et al. 2009).To explore the full parameter space and quantify degen-eracies, it uses an MCMC sampler. A Chabrier initialmass function (IMF) was assumed, but later, for com-parison with literature, converted to a Kroupa IMF (byadding 0.075 to log M ). This gives us stellar masses forthe different host-galaxy components of 63 objects; for3 objects, no robust stellar masses could be determined.Thus, the M BH -stellar-mass relations presented in sec-tion 4.2 include 63 objects.Given σ within the effective radius as described inthe previous section, we can also calculate a dynamicalmass: M sph , dyn = cr eff , sph σ , reff /G (1)with c = 3 for comparison with literature (Courteau etal. 2014). Since robust σ measurements within the effec-tive spheroid radius were only obtained for 50 objects,the M BH - M sph , dyn relation in section 4.2 includes 50 ob-jects. 3.4. Comparison samples
To compare the resulting scaling relations of M BH and σ , luminosity and mass with the literature, we use thesample presented by Kormendy & Ho (2013) as a qui-escent galaxy comparison sample, 85 local galaxies with M BH based on dynamical modeling of spatially-resolvedkinematics. Their sample consist of 44 elliptical galax-ies, 20 spiral and S0 galaxies with classical bulge and 21spiral and S0 galaxies with pseudo-bulge. Five of the el-liptical galaxies are mergers in progress. Pseudo-bulgesand mergers are significant outliers in Kormendy & Ho(2013) and ignored here. For 11 objects, the M BH is con-sidered uncertain and these objects are also ignored. Weare thus left with 51 objects total, 32 elliptical galaxiesand 19 spiral and S0 galaxies with classical bulges.The stellar-velocity dispersions are adopted in mostcases from G¨ultekin et al. (2009) and represent ef-fective velocity dispersions within r eff /2 as average of V ( r ) + σ ( r ) weighted by I ( r ) dr , thus consistent withthe way we derived stellar-velocity dispersions, sincethe difference between averaging inside r eff and r eff /2is small (Kormendy & Ho 2013).Kormendy & Ho (2013) list spheroid magnitudes in Ksand V, and (V-Ks) and (B-V) colors. We use a varietyof elliptical and spiral spectral templates from Bruzual& Charlot (2003) and Kinney et al. (1996) and derive alinear least-square fit of the form ( V − I ) = α ∗ ( B − V )+ β with α = 0 .
72 and β = 0 .
41, for conversion to I-bandmagnitudes.The stellar spheroid masses given by Kormendy & Ho(2013) are derived from a mean of mass-to-light ratios based on σ and ( B − V ) (their equations 8 and 9) and K-band magnitude. The mass-to-light ratio based on coloris derived from Into & Portinari (2013), who assume aKroupa (2001) IMF, but Kormendy & Ho (2013) adjustto the dynamical zeropoint.For the M BH - σ relation, we also show the 29 RM AGNsample presented by Woo et al. (2015). We adjust their M BH to match the virial factor of log f = 0 .
97 adoptedhere. RESULTS AND DISCUSSION4.1.
Host-galaxy morphology
The host-galaxy morphology was determined based onvisual inspection of images and the results of the surface-brightness fitting (Section 3.1). Of the full sample of 66AGNs with HST images, we conclude that 3 are hostedby bona-fide elliptical galaxies and 63 by spiral or S0galaxies. Out of the latter, 26 galaxies are found tohave a bar. In order for a spheroidal component to beclassified as pseudo-bulge, we conservatively require thatat least three of the following four criteria are met (fol-lowing Kormendy & Ho 2013): (i) S´ersic index <
2; (ii)bulge-to-total luminosity ratio < >
1; (iv) for face-on galaxies, the presenceof a bar is considered an indicator for the existence of apseudo-bulge. In this way, of the 63 spiral or S0 galaxies,21 spheroids are classified as pseudo-bulges, the major-ity of which (19) are in barred spiral galaxies. Table 1gives the host galaxy classification for all objects. Fourobjects show signs of interaction and/or merger activity(0206 − M BH -host-galaxy scaling relations with other parameters such as(pseudo) bulges and bars.4.2. Scaling relations
We present scaling relations between M BH and stel-lar velocity dispersion σ (within effective radius ofspheroid), dynamical spheroid mass, stellar mass, andI-band luminosity (Figure 5). We choose Kormendy &Ho (2013) for a consistent comparison for all these differ-ent scaling relations, even though there are more recentstudies with a compilation of larger samples. However,in a review by Greene et al. (2020), the authors notethat their results on the M BH - σ relation would not havechanged if, instead of using a recent literature compi- lack hole mass scaling relations for local active galaxies M BH - σ relation, we also show 29RM AGNs measured in a similar way (Woo et al. 2015,see Section 3.4 for details).Following common practices, we fit the different scal-ing relations as the linear relation with α and β asslope and intercept values (Table 3). The error barsof the measurements (in both x and y) are taken intoaccount to perform the inference; in this step we adoptthe EMCEE package (Python based) to derive the best-fit value and the uncertainties of α and β . The intrin-sic scatter is estimated so that when the squares of theobserved uncertainties are summed up, the best-fit re-duced chi-square value is close to unity. For all of the M BH scaling relations, our sample of 66 local AGNsnaturally extends the correlations for quiescent galax-ies down to M BH ∼ M (cid:12) along the same line, withthe same slope and normalization. However, by itself,the dynamic range in M BH covered by our sample is toosmall to determine the slope. Thus, when deriving fitsto the different scaling relations, we either fit both sam-ples (AGNs and quiescent galaxies) together or whenfitting our AGN sample alone, we fix the slope to thatof Kormendy & Ho (2013).The M BH - σ relation of our local AGN sample with M BH determined using the single-epoch method and σ based on spatially-resolved kinematics agrees well withthat of AGNs with M BH obtained from reverberationmapping (Park et al. 2012; Woo et al. 2015). The im-portance of the RM AGN sample cannot be overstatedsince it serves as the M BH calibrator beyond the localUniverse. Given that slope and scatter of the M BH - σ relation of our local AGNs, selected based solely onbroad H β emission line width, agree with that of RMAGNs provides an independent confirmation that theselection of the RM AGN sample based on variability(not on well defined galaxy/black-hole mass properties)does not introduce biases. Moreover, the close agree-ment between both samples provides an indirect vali-dation of the single-epoch method for the estimation of M BH . Note that these conclusions are independent ofthe fact that the M BH - σ scaling relation of RM AGNsis matched to that of inactive galaxies, since that onlyaffects the normalization, but not slope and scatter.To illustrate the effect of un-identified bars, we alsoinclude scaling relations for stellar mass and luminositywith spheroid+bar component added. This may helpin comparison with literature data, especially given thedifficulties and potential ambiguities involved in decom-position of images with poor data quality. Since thespheroids in the majority of barred spiral galaxies inour sample are classified as pseudo-bulges (19/21), this affects the location of pseudo-bulges the most. It movesthe pseudo-bulges further to the right in the M BH -stellarmass and M BH -luminosity relations which tends to movethem into better agreement with the scaling relations ofquiescent galaxies. However, within the uncertainty, thedifference is small. For none of the scaling relations dowe find a significant difference between pseudo- and clas-sical bulges in terms of correlations with M BH . This isin line with some studies (e.g., Davis et al. 2018), butcontrary to many others (e.g., Hu 2008; Greene et al.2010; Sani et al. 2011; L¨asker et al. 2016; Saglia et al.2016; Menci et al. 2016; de Nicola et al. 2019). For ex-ample, Kormendy & Ho (2013) went so far to concludethat “any M BH correlations with the properties of disk-grown pseudobulges [...] are weak enough to imply noclose coevolution” (see also, Kormendy et al. 2011).Pseudo-bulges, considered to have evolved secu-larly through dissipative processes rather than throughgalaxy mergers (e.g., Courteau et al. 1996; Kormendy &Kennicutt 2004), play an important role for understand-ing the origin of the M BH scaling relations. If majormergers are the fundamental drivers of the M BH scal-ing relations, only classical bulges, centrally concen-trated, mostly red and quiescent, merger-induced sys-tems, should follow these tight correlations. On the ba-sis of high-quality HST imaging, a careful analysis anda conservative classification of bulges as pseudo-bulges,our results clearly show that pseudo-bulges follow thesame relations as classical bulges, confirming findings ofan earlier study of ours based on SDSS images (Bennertet al. 2011a). This rules out hierarchical assembly asthe sole origin of the M BH -host-galaxy scaling relations(Peng 2007; Hirschmann et al. 2010; Jahnke & Mac-cio 2011; Kormendy & Ho 2013), a conclusion reachedby others, based on different arguments (e.g., Angl´es-Alc´azar et al. 2013; Graham 2016).In fact, our study shows that there are no significantoutliers that could be attributed to any specific category,whether it be galaxies with pseudo-bulges, bars or signsof interactions/mergers. For example, the four objectswith signs of mergers/interaction do not tend to lie offthe relations. Likewise, barred galaxies (26 out of 63 diskgalaxies in our sample) do not form outliers, in line withsome literature (Beifiori et al. 2012; Sahu et al. 2019).The location of barred galaxies on the scaling relationsis not only important since over half of the disk galaxypopulation is barred (e.g., Weinzirl et al. 2009), but alsobecause of the relevance of bars in secular evolution andpotentially fueling of BHs. Moreover, it is much easierto identify a bar than a pseudo-bulge (for a discussion,see Graham 2016), a reason why some studies chooseto distinguish between barred and non-barred galaxies6 Bennert et al. rather than classical vs. pseudo-bulges (e.g., Graham& Scott 2013). Most previous literature studies foundbarred galaxies to lie off the M BH -scaling relations, inparticular in the case of M BH - σ (e.g., Graham 2008;Graham & Li 2009; Hu 2008, for conflicting results, seealso Beifiori et al. 2012). This is not surprising, giventhat the stellar dynamics in galactic sub-structures suchas bars and pseudo-bulges is very different from that ofelliptical galaxies or classical bulges. Moreover, σ mea-surements can depend significantly on, e.g., size of thefiber (as is the case, for example, for SDSS), orientationof the slit (in case of long-slit observations) and aperturesize used (for details and comparisons, see Bennert et al.2015). Integral-field spectroscopy and spatially-resolvedspectroscopy is an obvious step forward and has been ob-tained for a sub-sample of the RM AGN sample (Batisteet al. 2017). While our σ measurements were obtainedusing long-slit spectroscopy, we mitigate these effects byusing spatially-resolved measurements integrated withinthe spheroid effective radius which is a robust way to de-termine σ (see also Bennert et al. 2015).The majority of AGNs in our sample reside in hostgalaxies of S0 or late-type morphology (63/66), out ofwhich almost half of the galaxies are barred and a thirdof spheroids are classified as pseudo-bulges. The factthat all of them are obeying the same tight M BH -scalingrelations, highlights the importance of secular evolutionfor the growth of BHs and bulges. Secular evolution mayhave a synchronizing effect, growing BHs and bulges si-multaneously at a small but steady rate for late-typegalaxies, and keeping them on tight relations over time.Comparison with semi-analytical models for galaxy for-mation including secular evolution (such as e.g., Menciet al. 2016, who, however, find little or no correlation ofpseudo-bulge mass with M BH ) can further shed light onsuch a scenario, but is beyond the scope of this paper.Interestingly, we do not find significant differences inthe tightness of the different correlations. The scatterin the relations ranges between 0.2-0.4 dex, smaller orequal to that of quiescent galaxies (G¨ultekin et al. 2009;McConnell & Ma 2013; Kormendy & Ho 2013). Thisis contrary to most previous studies that have oftenconcluded that M BH - σ is the tightest (0.3 dex in log M BH ) and thus the most fundamental of the scaling re-lations (Tremaine et al. 2002; Beifiori et al. 2012; Sagliaet al. 2016; Shankar et al. 2016; van den Bosch 2016; deNicola et al. 2019), at least for late-type spiral galaxies(G¨ultekin et al. 2009; Greene et al. 2010; L¨asker et al.2016, see, however, Davis et al. 2019). We attribute thisdifference to (i) our homogeneous sample selection, (ii)high-quality data, for both imaging and spectroscopy,and (iii) reliable surface photometry for a detailed struc- tural decomposition of the host galaxy components andspatially-resolved kinematics. Given the fact that thebiggest uncertainty in host-galaxy surface-brightness fit-ting is the classification and identification of individualstructures, a combination of (ii) and (iii) is essentialif one wants to determine the role of host-galaxy sub-structures on the correlation with M BH . SUMMARYThis paper presents a study of 66 local (0 . ≤ z ≤ .
1) active galactic nuclei (AGNs) homogeneously se-lected based on the presence of a broad H β emission linein SDSS spectra. High-quality HST optical (66 objects)and Gemini NIR imaging (14 of 66 objects) are com-plemented by spatially-resolved kinematics from spec-tra obtained at the Keck Telescopes. M BH is deter-mined based on the single-epoch method with broadH β emission-line width measured from Keck spectra.Surface photometry is performed using state of the artmethods, providing a structural decomposition of theAGN host galaxies into spheroid, disk and bar (whenpresent), with the spheroid component conservativelybeing classified as either classical or pseudo-bulge. Scal-ing relations between M BH -and host galaxy properties— spatially-resolved stellar-velocity dispersion, dynam-ical spheroid mass, stellar spheroid mass and spheroidluminosity — are presented, in comparison with quies-cent galaxies and RM AGNs taken from the literature.Our findings can be summarized as follows:1. The majority of AGNs (63/66) are hosted bygalaxies classified as spiral or S0 with a high frac-tion of bars (26/63) and pseudo-bulges (21/63),typical for Seyfert-type galaxies. The wide vari-ety of host-galaxy morphologies makes our sam-ple ideally suited to study the dependency of the M BH -host-galaxy scaling relations with other pa-rameters such as (pseudo) bulges and bars.2. Tight correlations are found between M BH andspatially-resolved stellar-velocity dispersion, dy-namical spheroid mass, stellar spheroid mass andspheroid luminosity, without significant differencesin the scatter. This is contrary to the widely ac-cepted paradigm that the M BH - σ relation is themost fundamental of all scaling relations.3. The intrinsic scatter of 0.2-0.4 dex is smallerthan or comparable to that of quiescent galaxies,showing that spiral galaxies hosting AGNs followthe same tight M BH -scaling relations, contrary tomany literature studies. lack hole mass scaling relations for local active galaxies M BH -host-galaxy scalingrelations and highlight the importance of secularevolution for growing both M BH and spheroid.5. The M BH - σ relation of our AGNs is indistinguish-able from the relation of AGNs with M BH ob-tained through reverberation mapping. This in-directly validates single-epoch virial estimators of M BH and is consistent with no significant selectionbias for RM AGNs.Our results show that all the tight correlations canbe simultaneously satisfied by AGN hosts in the 10 -10 M (cid:12) regime if data of sufficient quality are in hand andgreat care is taken when deriving host-galaxy proper-ties. A simple explanation of the difference between ouruniformly tight relations and the larger scatter found inthe literature is that σ is generally measured more ac-curately than the other host galaxy parameters. Thesample presented in this paper is meant to serve as alocal reference point for studies of the cosmic evolutionof the correlations between host galaxy properties and M BH . ACKNOWLEDGEMENTSWe thank Stephane Courteau, Alessandra Lamastra,Michael McDonald, Nicola Menci, Anowar Shajib, Dae-seong Park and Jong-Hak Woo for helpful discussions.VNB, IS and TS gratefully acknowledge assistance froma National Science Foundation (NSF) Research at Un-dergraduate Institutions (RUI) grant AST-1312296 andAST-1909297. Note that findings and conclusions donot necessarily represent views of the NSF. TT acknowl-edges support by NSF through grant AST-1907208, andby the Packard Foundation through a Packard ResearchFellowship. IS is supported by the German ResearchFoundation (DFG, German Research Foundation) underGermany’s Excellence Strategy - EXC 2121 “QuantumUniverse”- 390833306. Based on observations obtainedwith the Hubble Space Telescope and supported by aSpace Telescope Science Institute (STScI) grant associ-ated with program HST-GO-15215. Support for Pro-gram number HST-GO-15215 was provided by NASAthrough a grant from the Space Telescope Science In- stitute, which is operated by the Association of Univer-sities for Research in Astronomy, Incorporated, underNASA contract NAS5-26555 Based on observations ob-tained at the international Gemini Observatory, a pro-gram of NSF’s NOIRLab (processed using the GeminiIRAF package), which is managed by the Association ofUniversities for Research in Astronomy (AURA) under acooperative agreement with the National Science Foun-dation on behalf of the Gemini Observatory partnership:the National Science Foundation (United States), Na-tional Research Council (Canada), Agencia Nacional deInvestigaci´on y Desarrollo (Chile), Ministerio de Cien-cia, Tecnolog´ıa e Innovaci´on (Argentina), Minist´erio daCiˆencia, Tecnologia, Inova¸c˜oes e Comunica¸c˜oes (Brazil),and Korea Astronomy and Space Science Institute (Re-public of Korea). Based on observations obtained atthe W. M. Keck Observatory, which is operated as ascientific partnership among Caltech, the University ofCalifornia, and the National Aeronautics and Space Ad-ministration (NASA). The Observatory was made pos-sible by the generous financial support of the W. M.Keck Foundation. The authors recognize and acknowl-edge the very significant cultural role and reverence thatthe summit of Mauna Kea has always had within theindigenous Hawaiian community. We are most fortu-nate to have the opportunity to conduct observationsfrom this mountain. This research has made use ofthe Dirac computer cluster at the California PolytechnicState University in San Luis Obispo, maintained by Dr.Brian Granger and Dr. Ashley Ringer McDonald, andthe Hoffman2 Cluster at the University of California LosAngeles, managed and operated by the IDRE ResearchTechnology Group under the direction of Lisa Snyder.This research has made use of the public archive of theSloan Digital Sky Survey (SDSS) and the NASA/IPACExtragalactic Database (NED) which is operated by theJet Propulsion Laboratory, California Institute of Tech-nology, under contract with the National Aeronauticsand Space Administration. This research has made useof the NASA/IPAC Infrared Science Archive, which isfunded by the National Aeronautics and Space Adminis-tration and operated by the California Institute of Tech-nology. We thank Gemini staff observers K. Chibocas,W. Fraser, L. Fuhrmann, T. Geballe, M. Hoenig, J.Miller, S. Pakzad, R. Pike, M. Schwamb, O. Smirnova,and A. Smith for obtaining these data in queue mode. Facilities:
HST (WFC3), Keck:I (LRIS), Gem-ini:Gillett (NIRI), SloanAPPENDIX8
Bennert et al.
Figure 5. M BH scaling relations. In all panels, black data points correspond to the local quiescent comparison sample fromKormendy & Ho (2013), only including elliptical galaxies and spiral galaxies with classical bulge. For our sample, pseudo-bulgesare shown in blue and classical bulges in red. To reduce confusion of data points, error bars on M BH for our sample areomitted and shown instead in the bottom right corner. Top left panel: M BH - σ relation. Cyan data points show 29 RM AGNsfrom Woo et al. (2015). Top right panel: M BH - M sph , dyn relation. Middle left panel: M BH - M sph relation. Middle right panel: M BH - M sph+bar relation. Bottom left panel: M BH - L sph , I relation. Bottom right panel: M BH - L sph+bar , I relation. lack hole mass scaling relations for local active galaxies Table 3.
Fits to the Local Scaling Relations. X in relation Sample α β Scatter Reference(1) (2) (3) (4) (5) (6) σ/ − AGNs (50) & Quiescent galaxies (51) 8.52 ± ± ± σ/ − Quiescent galaxies (51) 8.49 ± ± ± σ/ − AGNs (50) 8.49 ± ± M sph , dyn / M (cid:12) AGNs (50) & Quiescent galaxies (52) 8.76 ± ± ± M sph , dyn / M (cid:12) AGNs (50) 8.87 ± ± M sph / M (cid:12) Quiescent galaxies (52) 8.69 ± ± M sph , stellar / M (cid:12) AGNs (63) & Quiescent galaxies (52) 8.72 ± ± ± M sph+bar , stellar / M (cid:12) AGNs (sph+bar) (63) & Quiescent galaxies (52) 8.71 ± ± ± M sph , stellar / M (cid:12) AGNs (63) 8.78 ± ± L sph , I / L (cid:12) AGNs (66) & Quiescent galaxies (51) 9.06 ± ± ± L sph , I / L (cid:12) AGNs (sph+bar) (66) & Quiescent galaxies (51) 9.05 ± ± ± L sph , I / L (cid:12) Quiescent galaxies (51) 9.11 ± ± ± L sph , I / L (cid:12) AGNs (66) 8.88 ± ± Note — The relations plotted as dashed lines in Fig. 5 correspond to the ones given in sample “AGN & Quiescent galaxies.” Col. (1):Scaling relation of the form log( M BH /M (cid:12) ) = α + β log X with X given in the table. Col. (2): Sample used for fitting. In parantheses,number of galaxies are given. Col. (3): Mean and uncertainty of the best fit intercept. Col. (4): Mean and uncertainty of the best fitslope. Col. (5): Mean and uncertainty of the best fit intrinsic scatter. Col. (6): References for fit. Kormendy & Ho (2013) is listed asKH13. Table 4 . Surface-Photometry Fitting Results.
HST I -band Gemini Ks -band HSTObject AGN Spheroid Disk Bar AGN Spheroid Disk Bar n sph R sph R disk R bar (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) )(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)0013-0951 18.3 18.0 15.9 ... 16.0 17.5 15.0 ... 1.2 0.36 5.19 ...0038+0034 17.9 16.3 18.3 ... ... ... ... ... 4.5 2.39 2.39 ...0109+0059 18.9 18.3 16.9 18.2 17.8 17.1 15.9 17.2 1.6 0.2 3.2 1.110121-0102 16.7 18.4 14.8 16.6 15.1 17.0 14.1 15.6 1.0 0.42 5.35 2.710150+0057 19.1 17.6 15.5 17.2 17.0 16.8 14.6 16.2 1.0 0.39 3.92 1.30206-0017 19.0 14.1 14.2 ... 16.0 13.1 13.5 ... 3.8 3.27 9.42 ...0212+1406 18.4 16.8 15.6 18.4 ... ... ... ... 1.0 0.75 4.22 0.370301+0110 18.1 16.9 17.7 ... ... ... ... ... 4.0 1.2 3.35 ...0301+0115 18.0 18.1 16.8 18.1 15.9 16.4 15.7 17.3 1.0 0.23 2.95 1.460336-0706 20.7 17.2 16.4 ... ... ... ... ... 1.0 0.77 5.4 ...0353-0623 18.4 17.3 16.9 18.4 ... ... ... ... 1.0 0.94 4.75 0.270737+4244 19.7 17.5 16.6 ... ... ... ... ... 2.7 0.11 2.85 ...0802+3104 17.2 17.4 15.6 17.7 ... ... ... ... 1.1 0.28 3.27 1.010811+1739 19.1 17.7 16.2 17.9 ... ... ... ... 1.4 0.38 4.48 2.880813+4608 20.1 16.6 16.6 17.0 17.8 15.7 15.7 16.1 2.6 0.67 5.14 3.580845+3409 20.1 16.9 15.9 18.1 17.4 15.7 15.2 17.2 3.9 1.09 7.0 1.660857+0528 18.0 17.7 15.6 ... ... ... ... ... 1.2 0.38 3.56 ...0904+5536 16.7 16.4 16.6 ... ... ... ... ... 1.8 1.35 7.42 ...0909+1330 20.6 17.7 15.4 17.3 ... ... ... ... 1.5 0.64 7.61 4.61 Table 4 continued Bennert et al.
Table 4 (continued)
HST I -band Gemini Ks -band HSTObject AGN Spheroid Disk Bar AGN Spheroid Disk Bar n sph R sph R disk R bar (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) )(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)0921+1017 18.7 15.7 14.7 ... ... ... ... ... 5.0 3.88 3.97 ...0923+2254 16.4 15.5 14.2 16.4 ... ... ... ... 1.4 0.95 10.9 6.40923+2946 21.1 15.5 ... ... ... ... ... ... 4.8 2.37 ... ...0927+2301 17.7 14.8 13.5 ... ... ... ... ... 1.4 1.24 7.46 ...0932+0233 18.3 16.9 16.2 ... ... ... ... ... 1.1 0.68 4.37 ...0936+1014 16.9 18.0 15.2 ... ... ... ... ... 1.4 0.65 6.15 ...1029+1408 18.3 15.4 16.6 ... ... ... ... ... 4.3 2.42 5.87 ...1029+2728 19.8 16.2 16.5 ... ... ... ... ... 3.1 0.78 3.21 ...1029+4019 17.4 17.7 16.2 ... ... ... ... ... 1.1 0.42 2.5 ...1042+0414 18.8 17.6 16.5 18.1 ... ... ... ... 1.6 0.25 3.21 2.021043+1105 16.9 16.8 ... ... ... ... ... ... 3.1 2.3 ... ...1058+5259 18.3 16.9 16.7 17.3 ... ... ... ... 1.9 0.63 5.94 2.581101+1102 19.1 15.4 15.7 ... ... ... ... ... 5.0 3.0 4.59 ...1104+4334 20.6 15.8 18.7 18.6 ... ... ... ... 4.5 2.79 2.79 1.461137+4826 22.9 17.5 17.5 ... ... ... ... ... 2.6 0.24 1.29 ...1143+5941 18.7 17.5 16.5 17.5 ... ... ... ... 2.4 0.5 6.72 4.261144+3653 16.3 15.8 15.0 ... ... ... ... ... 1.4 1.2 7.87 ...1145+5547 19.3 18.2 15.3 18.0 ... ... ... ... 1.0 0.52 7.0 2.11147+0902 16.6 15.9 18.2 ... ... ... ... ... 3.8 2.01 2.01 ...1205+4959 17.2 16.6 15.7 ... ... ... ... ... 2.0 0.68 4.41 ...1206+4244 17.4 17.2 15.5 17.1 ... ... ... ... 1.0 0.64 7.0 2.571216+5049 18.4 15.0 14.9 ... ... ... ... ... 3.3 2.75 8.03 ...1223+0240 16.6 15.3 15.0 ... ... ... ... ... 5.0 3.51 4.3 ...1246+5134 19.1 17.9 17.1 ... ... ... ... ... 2.0 0.4 2.88 ...1306+4552 20.9 18.3 15.6 17.1 ... ... ... ... 1.0 0.3 5.05 2.881307+0952 19.1 17.4 15.3 19.4 ... ... ... ... 1.0 0.56 4.59 0.271312+2628 17.5 17.9 15.5 17.9 ... ... ... ... 1.0 0.45 6.03 2.71405-0259 18.8 17.4 15.6 ... ... ... ... ... 1.0 0.77 5.01 ...1416+0137 18.5 15.9 15.2 ... ... ... ... ... 2.4 1.57 6.75 ...1419+0754 17.6 16.1 14.5 ... ... ... ... ... 1.9 0.98 6.45 ...1434+4839 17.7 16.3 15.0 16.4 ... ... ... ... 1.5 0.88 6.29 3.981545+1709 18.7 16.2 16.9 ... ... ... ... ... 5.0 1.49 2.42 ...1557+0830 17.9 16.8 ... ... ... ... ... ... 2.5 1.16 ... ...1605+3305 17.9 16.9 16.5 ... ... ... ... ... 1.2 0.68 3.82 ...1606+3324 18.6 16.3 16.5 ... ... ... ... ... 3.3 1.16 4.29 ...1611+5211 18.8 15.8 16.2 ... ... ... ... ... 3.1 0.75 4.62 ...1636+4202 18.4 15.8 17.1 18.2 ... ... ... ... 5.0 3.0 3.0 0.331708+2153 16.6 16.5 16.3 ... ... ... ... ... 1.9 1.54 6.95 ...2116+1102 18.1 17.6 16.0 19.0 ... ... ... ... 1.2 0.5 5.95 1.962140+0025 16.5 17.3 16.8 ... 15.0 17.0 15.7 ... 1.0 0.47 2.24 ...2215-0036 17.0 18.1 16.7 ... ... ... ... ... 1.0 0.5 3.45 ...2221-0906 17.6 17.6 17.3 ... 16.6 17.6 15.9 ... 2.7 0.86 2.98 ...2222-0819 17.4 18.7 15.6 17.8 15.0 25.9 14.7 16.6 1.1 0.33 3.98 1.482233+1312 17.9 17.9 15.9 17.5 16.2 16.4 14.8 16.2 1.0 0.38 7.71 1.82 Table 4 continued lack hole mass scaling relations for local active galaxies Table 4 (continued)
HST I -band Gemini Ks -band HSTObject AGN Spheroid Disk Bar AGN Spheroid Disk Bar n sph R sph R disk R bar (mag) (mag) (mag) (mag) (mag) (mag) (mag) (mag) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) ) ( (cid:48)(cid:48) )(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)2254+0046 17.0 17.4 17.6 ... ... ... ... ... 1.1 0.26 1.91 ...2327+1524 17.9 14.7 14.8 ... 15.1 13.5 14.1 ... 2.4 1.71 7.86 ...2351+1552 17.7 17.0 16.9 ... 16.5 16.1 15.8 ... 2.0 0.77 2.69 ... Note —Surface-photometry fitting results using lenstronomy on HST and Gemini images. Col. (1): Target ID used throughoutthe text (based on R.A. and declination). Col. (2): Point-source (AGN) magnitude in HST I -band (uncertainty 0.1 mag). Col.(3): Spheroid magnitude in HST I -band (uncertainty 0.1 mag). Col. (4): Disk magnitude in HST I -band (if present; uncertainty0.1 mag). Col. (5): Bar magnitude in HST I -band (if present; uncertainty 0.1 mag). Col. (6): Point-source (AGN) magnitudein Gemini Ks -band (uncertainty 0.2 mag). Col. (7): Spheroid magnitude in Gemini Ks -band (uncertainty 0.2 mag). Col. (8):Disk magnitude in Gemini Ks -band (if present; uncertainty 0.2 mag). Col. (9): Bar magnitude in Gemini Ks -band (if present;uncertainty 0.2 mag). Col. (10): Spheroid S´ersic index n (5% uncertainty). Col. (11): Spheroid radius in arcseconds (10%uncertainty). Col. (12): Disk radius in arcseconds (10% uncertainty). Col. (13): Bar radius in arcseconds (10% uncertainty). REFERENCES
Abazajian, K. N., Adelman-McCarthy, J. K., Agueros, M.A., et al. 2009, ApJS, 182, 543Adelman-McCarthy, J. K., Agueros, M. A., Allam, SaharS., et al. 2008, ApJS, 175, 297Angl´es-Alc´azar, D., ¨Ozel, F., & Dav’e, R. 2013, ApJ, 770, 5Auger, M. W., Treu, T., Bolton, A. S., et al. 2009, ApJ,705, 1099Batiste, M., Bentz, M. C., Raimundo, S. I., Vestergaard,M., & Onken, C. A. 2017, ApJL, 838, 10Beifiori, A., Courteau, S., Corsini, E. M., & Zhu, Y. 2012,MNRAS, 419, 2497Bennert, V. N., Treu, T., Woo, J.-H., et al. 2010, ApJ, 708,1507Bennert, V. N., Auger, M. W., Treu, T., Woo, J.-H., &Malkan, M. A. 2011a, ApJ, 726, 59Bennert, V. N., Auger, M. W., Treu, T., Woo, J.-H., &Malkan, M. A. 2011b, ApJ, 742, 107Bennert, V. N., Treu, T., Auger, M. W., et al. 2015, ApJ,809, 20Birrer, S., Amara, A., & Refregier, A. 2015, ApJ, 813, 102Birrer, S. & Amara, A. 2018, Physics of the Dark Universe,22, 189Brewer, B. J., Treu, T., Pancoast, A., et al. 2011, ApJL,733, L33Bruzual, G., & Charlot, S. 2003, MNRAS, 344, 1000Cisternas, M, Jahnke, K., Bongiorno, A. et al. 2011a, ApJ,741, 11Cisternas, M, Jahnke, K., Inskip, K. et al. 2011a, ApJ, 726,57Courteau, S., de Jong, R. S., & Broeills, A. H. 1996, ApJL,457, 73 Courteau, S., Cappellari, M., de Jong, R. S., et al. 2014,Reviews of Modern Physics, 86, 47Croton, D. J. 2006, MNRAS, 369, 1808Davis, B. L., Graham, A. W., & Cameron, E. 2018, ApJ,869, 113Davis, B. L., Graham, A. W., & Cameron, E. 2019, ApJ,873, 85Decarli, R., Falomo, R., Treves, A., et al. 2010, MNRAS,402, 2453DeGraf, C., Di Matteo, T., Treu, T., et al. 2015, MNRAS,454, 913de Nicola, S., Marconi, A., & Longo, G. 2019, MNRAS,490, 600Ding, X., Sliverman, J., Treu, T., et al. 2020, ApJ, 888, 37Di Matteo, T. Springel, V. & Hernquist, L. 2005, Nature,433, 604Dubois, Y., Gavazzi, R., Peirani, S. & Silk, J. 2013,MNRAS, 433, 3297Dubois, Y., Peirani, S., Pichon, C., et al. 2016, MNRAS,463, 3948Ferrarese, L., & Ford, H. 2005, Space Science Reviews, 116,523Graham, A. W. 2008, ApJ, 680, 143Graham, A. W. 2016, Astrophysics and Space ScienceLibrary, 418, 263Graham, A. W., & Scott, N. 2013, ApJ, 764, 151Graham, A. W., Li, & I. -hui 2009, ApJ, 698 812Greene, J. E., & Ho, L. C. 2006, ApJL, 641, 21Greene, J. E., Peng, C. Y., Kim, M., et al. 2010, ApJ, 721,26Greene, J. E., Strader, J., & Ho, L. 2020, ARA&A, 58, 257 Bennert et al.
G¨ultekin, K., Richstone, D. O., Gebhardt, K., et al. 2009,ApJ, 698, 198Harris, W. E. 2018, AJ, 156, 296Harris, C. E., Bennert, V. N., Auger, M. W., et al. 2012,ApJS, 201, 29Hirschmann, M., Khochfar, S., Burkert, A., et al. 2010,MNRAS, 407, 1016Hopkins, P. F., Hernquist, L., Cox, T. J., Robertson, B., &Krause, E. 2007, ApJ, 669, 45Hopkins, P. F., Torrey, P., Faucher-Gigu´ere, C.-A.,Quataert, E., & Murray, N. 2016, MNRAS, 458, 816Hu, J., 2008, MNRAS, 386, 2242Into, T., & Portinari, L. 2013, MNRAS, 430, 2715Jahnke, K. & Maccio, A. 2011, ApJ, 734, 92Jahnke, K., Bongiorno, A., Brusa, M., et al. 2009, ApJL,706, 215Kennedy, J. & Eberhart, R. C. 2001, Academic PressKinney, A. L., Calzetti, D., Bohlin, R. C., et al. 1996, ApJ,467, 38Kormendy, J. & Ho, L. C. 2013, ARA&A 51, 511Kormendy, J. & Kennicutt 2004, ARA&A, 42, 604Kormendy, J., Bender, R., & Cornell, M. E. 2011, Nature,469, 374L¨asker, R., Greene, J. E., Seth, A., et al. 2016, ApJ, 825, 1Li, Y.-R., Wang, J.-M., Ho, L. C., Du, P., & Bai, J.-M.2013, ApJ, 779, 110McConnell, N. J. & Ma, C.-P., 2013, ApJ, 764, 184McConnell, N. J., Ma, C.-P., Gebhardt, K. et al. 2011,Nature, 480, 215McGill, K. L., Woo, J.-H., Treu, T., & Malkan, M. A. 2008,ApJ, 673, 703Menci, N., Fiore, F., Bongiorno, A., & Lamastra, A. 2016,˚a, 594, 99Merloni, A., Bongiorno, A., Bolzonella, M., et al. 2010,ApJ, 708, 137Mortlock, D., Warren, S. J., Venemans, B.P., et al. 2011,Nature, 474, 616Onken, C. A., Ferrarese, L., Merritt, D., et al. 2004, ApJ,615, 645Pancoast, A., Brewer, B. J., & Treu, T. 2011, ApJ, 730, 139Pancoast, A., Barth, A.J., Horne, K., et al. 2018, ApJ, 856,108Park, D., Woo, J. H., Treu, T., et al. 2012 ApJ, 747, 30Park, D., Woo, J.-H., Bennert, V. N., et al. 2015, ApJ, 799,164Peng, C. Y. 2007, ApJ, 671, 1098Peng, C., Ho, L. C., Impey, C. D. & Rix, H.-W. 2002, AJ,124, 266Peng, C. Y., Impey, C. D., Ho, L. C., Barton, E. J., & Rix,H.-W. 2006a, ApJ, 640, 114 Peng, C. Y., Impey, C. D., Rix, H.-W., et al. 2006b, ApJ,649, 616Prsa, A., Harmanec, P. Torres, G., et al. 2016, AJ, 152, 41Rakshit, S., Stalin, C. S., & Kotilainen, J. 2020, ApJS, 249,17Riechers, D. A., Walter, F., Carilli, C. L., & Lewis, G. F.2009, ApJ, 690, 463Runco, J., Cosens, M., Bennert, V. N., et al. 2016, ApJ,821, 33Saglia, R. P., Opitsch, M., Erwin, P., et al. 2016, ApJ, 818,47Salviander, S., Shields, G. A., Gebhardt, K. & Bonning E.W. 2007, ApJ, 662, 131Sahu, N., Graham, A. W., & Davis, B. L., ApJ, 887, 10Sani, E., Marconi, A., Hunt, L.K., & Risaliti, G. 2011,MNRAS, 413, 1479Schlafly, E. F. & Finkbeiner, D. P. 2011, ApJ 737, 103S´ersic J. L., 1963, Boletin de la Asociacion Argentina deAstronomia La Plata Argentina, 6, 41Sexton, R. O., Canalizo, G., Hiner, K.D., et al. 2019, ApJ,878, 101Shankar, F., Bernardi, M., Sheth, R.K., et al. 2016,MNRAS, 460, 3119Shankar, F., Bernardi, M., Richardson, K., et al. 2019,MNRAS, 485, 1278Shen, Y. 2013, BASI, 41, 61Silverman, J. D., Treu, T., Ding, X., et al. 2019, ApJL, 887,5Tremaine, S., Gebhardt, K., Bender, R., et al. 2002, ApJ,574, 740Treu, T., Malkan, M. A., & Blanford, R. D. 2004, ApJL,615, 97van den Bosch, R. C. 2016, ApJ, 831, 134van Dokkum, P. 2001, PASP, 113, 1420)Vestergaard, M. 2002, ApJ, 571, 733Vestergaard, M. & Peterson, B. M. 2006, ApJ, 641, 689Wandel, A., Peterson, B. M., & Malkan, M. A. 1999, ApJ,526, 579Weinzirl, T., Jogee, S., Kochfar, S., et al. 2009, ApJ, 696,411Williams, P. R., Pancoast, A., Treu, T., et al. 2018, ApJ,866, 75Williams, P. R., Pancoast, A., Treu, T., et al. 2020, ApJ,902, 74Willmer, C. N. A. 2018, ApJS, 236, 47Woo, J.-H., & Urry, C. M. 2002, ApJ, 579, 530Woo, J.-H., Treu, T., Malkan, M. A., & Blandford, R. D.2006, ApJ, 645, 900Woo, J.-H., Treu, T., Barth, A. J., et al. 2010, ApJ, 716,269 lack hole mass scaling relations for local active galaxies23