Fading AGN candidates: AGN histories and outflow signatures

19 

Full text

(1)

© 2017. The American Astronomical Society. All rights reserved.

Fading

AGN

Candidates:

AGN

Histories

and

Out

ow

Signatures

WilliamC.Keel1,ChrisJ. Lintott2,W.Peter Maksym1,3,VardhaN.Bennert4,S.DrewChojnowski5,AlexeiMoiseev6, AleksandrinaSmirnova6,KevinSchawinski7,LiaF.Sartori7,C.MeganUrry8,AnnaPancoast3,10,MischaSchirmer9,BryanScott4,

CharlesShowley4,andKelsiFlatland4

1

DepartmentofPhysicsandAstronomy,UniversityofAlabama,Box870324,Tuscaloosa,AL35487,USA; wkeel@ua.edu 2

Astrophysics,OxfordUniversity;andAdlerPlanetarium,1300S.LakeshoreDrive,Chicago,IL60605,USA

3

CenterforAstrophysics,60GardenSt.,Cambridge,MA02138,USA

4 PhysicsDepartment,CaliforniaPolytechnicStateUniversity,SanLuisObispo,CA93407,USA 5

DepartmentofAstronomy,NewMexicoStateUniversity,P.O.Box30001,MSC4500,LasCruces,NM88003-8001,USA

6

SpecialAstrophysicalObservatory,RussianAcademyofSciences,NizhnyArkhyz,369167,Russia

7

InstituteforAstronomy,ETHZürich,Wolfgang-Pauli-Straße27,CH-8093Zurich,Switzerland 8

DepartmentofPhysics,YaleUniversity,P.O.Box208120,NewHaven,CT06520-8120,USA

9

GeminiObservatory,LaSerena,Chile

Received2016August31;revised2016December2;accepted2016December18;published2017February1

Abstract

Weconsidertheenergybudgetsandradiativehistoryofeightfadingactivegalacticnuclei(AGNs),identifiedfrom an energy shortfallbetween the requirementstoionize very extended(radius>10kpc)ionizedclouds and the luminosity of the nucleus as we view it directly. All show evidence of significant fading on timescales of

≈50,000yr.WeexploretheuseofminimumionizingluminosityQionderivedfromphotoionizationbalanceinthe brightestpixelsinHαateachprojectedradius.TestsusingpresumablyconstantPalomar–GreenQSOs,andoneof ourtargetswithdetailedphotoionizationmodeling,suggestthatwecanderiveusefulhistoriesofindividualAGNs, withthecaveatthattheminimumionizingluminosityisalwaysanunderestimateandsubjecttouncertaintiesabout

finestructureintheionizedmaterial.Theseconsistencytestssuggestthatthedegreeofunderestimationfromthe upperenvelopeofreconstructedQionvaluesisroughlyconstantforagivenobjectandthereforedoesnotprevent suchderivation.TheAGNsinoursampleshowarangeofbehaviors,withrapiddropsandstandstills;thecommon featureisarapiddropinthelast≈2×104yrbeforethedirectviewofthenucleus.Thee-foldingtimescalesfor ionizingluminosityaremostly inthethousands ofyears,withafewepisodesas shortas 400yr.In thelimitof largelyobscuredAGNs,we findadditionalevidenceforfadingfromtheshortfallbetween eventhelowerlimits fromrecombination balanceand themaximum luminositiesderived fromfar-infrared fluxes.Wecomparethese long-term light curves, and the occurrence of these fading objects among all optically identified AGNs, to simulations ofAGNaccretion;the strongestvariationsover thesetimespans areseeninmodelswithstrongand local(parsec-scale)feedback.WepresentGeminiintegral-fieldopticalspectroscopy,whichshowsaverylimited role foroutflowsinthese ionizedstructures. Whileringsand loops of emission, morphologicallysuggestiveof outflow, arecommon, theirkinematic structure shows sometobe inregularrotation.UGC 7342 exhibitslocal signaturesofoutflows<300kms−1,largelyassociatedwithverydiffuseemission,andpossiblyentraininggasin oneofthecloudsseeninHubbleSpaceTelescopeimages.OnlyintheTeacupAGNdoweseeoutflowsignatures of theorder of 1000kms−1.In contrastto the extendedemission regionsaround many radio-loud AGNs, the cloudsaroundthesefadingAGNsconsistlargelyoftidaldebrisbeingexternallyilluminatedbutnotdisplacedby AGNoutflows.

Keywords:galaxies:active–galaxies:individual(NGC5792,UGC 7342,Mkn1498)– galaxies:interactions– galaxies: Seyfert

1. Introduction spiralgalaxyIC2497.Knownafteritsdiscoverer asHanny’s

Voorwerp, this object shows high-ionization emission lines It has long been known that some active galactic nuclei

withratiosessentiallyidenticaltothenarrow-lineregionofan

(AGNs) are accompanied by extended emission-line regions

AGN,spanningaprojected rangefrom15to35kpc fromthe

(EELRs),zonesofionizedgasspanninggalaxyscalesoreven

galaxynucleus,whichfailsbyatleasttwoordersofmagnitude larger. Such regions can trace the geometry of ionizing

inbolometric output tomatch the ionization requirements of radiation escaping the AGN and host galaxy, and at least

thecloud(Lintottetal.2009; Keeletal. 2012b).Thecloud’s implicitly give hints to the luminosity history of the AGN.

electron temperature, narrow line widths, and quiescent EELRsoccuraroundSeyfertnuclei,QSOs,andradiogalaxies.

velocity field indicate photoionization rather than shocks as A very luminous and extensive EELR was found in the

theenergy source. HI mappingby Józsaetal. (2009)shows courseoftheGalaxyZooproject(Lintottetal.2008)nearthe

thistobetheionizedpartofa300kpctrailofotherwiseneutral gas,suggestingastronggalaxyinteractiongigayearsago.The

Based on observations with the NASA/ESA Hubble Space Telescope energy mismatch between the nucleus of IC 2497 and the

obtainedat theSpaceTelescopeScienceInstitute,whichisoperatedbythe ionizationrequirementsinHannysVoorwerpindicatesthatthe Association ofUniversities forResearch in Astronomy, Inc., under NASA

contractNo.NAS5-26555. nucleus faded from a luminosity associated with QSOs to a

10

(2)

StartingwiththeunusualmorphologyandSloanDigitalSky Survey (SDSS) gri colors of Hanny’s Voorwerp, a targeted searchbyGalaxyZoovolunteersfound19additionalEELRs, including hithertounknownexamples,ofwhich eightshowed energyshortfallssimilartothatinIC2497(Keeletal.2012a). This sample was selected in a homogeneous way, requiring detection of emission features showing AGN-like line ratios more than 10kpc in projection from the nucleus. We have pursuedadetailedstudyofthissubsetofeight(plusIC2497/ Hanny’sVoorwerp)tolearnmoreaboutAGNvariationsonthe otherwise inaccessible range of timescales of 104–105yr. Analogous cases of extended ionized gas near energetically inadequate AGNs have been reported at higher redshift and luminositybySchirmeretal.(2013,2016),their“GreenBean” systems, and at low luminosity in the local universe in the mergerremnantNGC7252(Schweizeretal.2013)andasone interpretation of an off-nuclear ionized region in the spiral galaxyNGC3621(Menezesetal.2016).

Interest inthe historyof accretiononto nuclearblack holes has alsobeenrenewed byevidencefor aneventful history of outburstswithin106yrbythecurrentlyquiescentblackholeat theGalacticCenter,fromX-rayechoes(Munoetal.2007),the ionizationstructureoftheMagellanicStream(Bland-Hawthorn et al. 2013), and possibly the “Fermi bubbles” (Zubovas etal.2011;Guo&Mathews 2012).

The variations in AGN luminosity we infer from these objects, spanningscales upto 105yr, connect implicitly with the variations on timescales we canobserve directly. “Chan­ ging-look” AGNs can have broad-line emission essentially vanishover≈10yrintheemittedframe(MacLeodetal.2015; Ruanetal.2015;Runcoetal.2016;Runnoeetal.2016),while the narrow-line region is often so large that light-travel-time smearing leaves its emission nearly constant; similar inter­ pretationswerediscussedinthecontextofspectralvariationsin SeyfertnucleiatleastasearlyasPenston&Perez(1984).The temporal spectrum of variations inAGN outputclearly spans many ordersof magnitude, and affects ourunderstanding not onlyoftheirphysicalstructurebutalsoofthedemographicsof accretioninthegalaxypopulation.Seeingdimepisodessolong that theyappear insurroundingionizedgaswhen smearedby therecombinationtimescalesofthousands ofyears,aswell as geometricprojectionfactors,meanseither thattheAGNsstay inaverylowstatefortheselong timesorthatexcursionstoa high statefillonlyasmallfractionofthetime.Understanding this history is key to understanding the broad idea of duty cyclesinAGNaccretion;X-raysurveysinparticularsuggesta dutycycleconnectedtotheEddingtonlimitandevolvingwith cosmic time (e.g., Shankar et al. 2009), but not how many episodesof whatdurationsareinvolved.

In a similarsense, radio observations suggest very episodic production of jets, a major form of kinetic energy output from accretion. The identification of radio jets and a circum­ nuclear outflow in IC 2497 (Józsa et al. 2009; Rampadarath etal.2010; Keel etal.2012b)ledustospeculatethatsomeofthe extremevariabilityneededtoexplaintheionizationofHanny’s Voorwerpmightcomenotsolelyfrom adropinthe accretion-drivenluminosityoftheAGNinIC2497,butfromsomeofthe “missing”luminosity switchingto akineticmode. Thisideais supportedbya“bubble”ofhotgasaroundthefadingAGNofIC 2497seenwithChandra(Sartorietal.2016).

All these factors motivated more detailed study of our fading-AGN candidates with avariety of techniques. Paper I

(Keeletal.2015)presentedresultsofHubbleSpaceTelescope

(HST)imagingconcerningthehost galaxiesand theorigin of theionizedgas.Thehostsarebulge-dominated,andeveryone shows signatures of ongoing or past galaxy interactions, 1.5–3Gyragoforsystemswithfavorablegeometrytoestimate this timescale. The extended gas has modestly subsolar metallicity,and is kinematically rotation-dominated; outflows have only a very localized role, in contrast to most QSO EELRs(Stocktonetal.2006).Theseareexternallyilluminated tidal debris; merger remnants are particularly good environ­ mentsforHItooccurfarfromthenucleusandoutofanyhost diskplane,and act asascreentoshow the escapingionizing radiation.Thisappearstobesuchastrongselectionfactorthat we donotknow whetherhigh-amplitude variability onscales of104yrisconfined tomergeraftermathsor not.

Inthispaper,wedescribeadditionaldata,includingGemini integral-field spectroscopy, and present implications of the entiredatasetforthepropertiesandhistoryoftheseAGNs.We use the peak surface brightness in recombination lines at variousprojectedradiitoreconstruct theluminosityhistoryof theAGNs,comparestructuresseenintheemission-lineimages and line ratios to address the role of ionization cones, and examine the occurrence of outflows near the nuclei to trace the role and extent of feedback near the epochs of radiative dimming.

In evaluatingsizesand luminosities, we computedistances usingaHubbleconstant of73kms−1Mpc−1.

2.Observations

2.1.GalaxySample

As given by Keel et al. (2015), the sample of AGNs we considerincludestheeightsystemsfromtheGalaxyZoosurvey of giant ionized clouds (Keel et al. 2012a) where a simple estimateofenergybudgetincludingtheionizationrequirements of the most distant emission-line region, compared to the observedbolometricluminosityofthenuclei,showedashortfall greaterthanaboutafactor5.TheseincludeNGC5252withits well-documented set of ionization cones (Prieto & Freud-ling1996;Morseetal.1998),andthelarge-scaledoubleradio sourcehostsNGC5972(Véron-Cetty&Véron2001)andMkn 1498(Röttgeringetal.1996).Additionalobjects,someshowing larger and more luminous EELRs, were newly found in this survey. This work deals with the ≈40% of the Galaxy Zoo EELRobjectsshowingevidenceforAGNfading;theremainder haveenergybudgets indicating roughly constant luminosityin somecaseswithsubstantialobscurationalongourlineofsight. Overall properties of these galaxies are listed in Table 1, including IC 2497 (host galaxy of Hanny’s Voorwerp) for reference.

2.2.HST Imaging

Centraltoouranalysisisasetofemission-lineimagesofour fading-AGNcandidatesobtainedwiththeHSTandnarrowband

filters inthe Wide Field Camera 3 (WFC3) or tunable ramp

filters on the Advanced Camera for Surveys (ACS), using archival WFPC2 data for NGC 5252. These data and our processing are described in detail by Keel et al. (2015). For brevity, we truncate the SDSS designations of two systems without more common catalog names, SDSS J151004.01

(3)

Table1

GalaxyProperties

Galaxy SDSSName Type z Scale(kpcarcsec–1)

IC2497 SDSSJ094104.11+344358.4 2 0.0502 0.96

Mkn1498 SDSSJ162804.06+514631.4 1.9 0.0547 1.11

NGC5252 SDSSJ133815.86+043233.3 1.5 0.0228 0.50

NGC5972 SDSSJ153854.16+170134.2 2 0.0297 0.63

SDSS1510+07 SDSSJ151004.01+074037.1 2 0.0458 0.89

SDSS2201+11 SDSSJ220141.64+115124.3 2 0.0296 0.59

TeacupAGN SDSSJ143029.88+133912.0 2 0.0852 1.58

UGC7342 SDSSJ121819.30+291513.0 2 0.0477 0.98

UGC11185 SDSSJ181611.61+423937.3 2 0.0412 0.80

Table2

WISEMagnitudesandIRLuminosities

Galaxy W1 (3.4μm) W2 (4.6μm) W3 (12μm) W4 (22μm) LogL(MIR) LogL(FIR)

IC2497 11.46±0.02 11.15±0.02 7.31±0.02 4.54±0.03 44.28 44.77

Mkn1498 10.26±0.02 9.24±0.02 6.31±0.02 3.72±0.02 45.41 <43.70

NGC5252 9.19±0.02 8.38±0.02 6.35±0.01 4.47±0.02 44.28 43.60

NGC5972 11.02±0.02 10.49±0.02 7.05±0.02 4.61±0.02 44.23 <43.74

SDSS1510+07 12.93±0.04 12.81±0.03 11.46±0.12 >9.00 43.15 <44.60

SDSS2201+11 11.10±0.02 11.08±0.02 8.60±0.02 6.77±0.07 43.57 <43.78

TeacupAGN 11.67±0.02 10.50±0.02 7.14±0.01 4.16±0.02 45.42 <43.36

UGC7342 12.70±0.02 12.61±0.03 9.90±0.05 7.17±0.10 43.70 <44.04

UGC11185 11.22±0.02 10.61±0.02 7.85±0.02 4.88±0.02 44.53 <44.25

Note.WISEdataaremagnitudesinfiltersW1–W4.Luminositiesarecomputedfortheranges3.4–42μmforMIRand42–122μmforFIR,expressedasthedecimal logofthevaluesinergs−1.

2.3.WISEData

OurestimatesofthebolometricluminosityoftheAGNs,and hencetheeffectofobscurationontheionizingradiation,canbe updatedfromKeeletal.(2012a)usingthemoresensitiveand higher-resolution survey data from the Wide-field Infrared SurveyExplorer(WISE,Wrightetal.2010)inplaceofIRASor

Akariforthemid-infraredregionincluding22μm.WISEdata show thatmuchofthemid- andfar-IRfluxattributedinitially toUGC 7342atz=0.047comesinsteadfromabackground starburst system at z =0.069 (as identified inPaper I). The AGN luminositywe derivefor UGC7342 iscorrespondingly lower, and the energy deficit for ionizing its gas filaments becomesevengreater.

The WISE catalog magnitudes are listed in Table 2. For furtheruse,weconvertedthemintofluxesusingthezeropoints fromCutrietal.(2014)for“compromise”spectralslopes.

The WISE dataare especially important since much of the reradiation from circumnuclear dust occurs in the mid-IR, which waspoorly covered by previous surveys.The infrared output of these galaxies, and thus the potential fraction of obscured and reradiated AGN radiation, can be quantified in severalways.Wehavederivedtotalfluxesandluminositiesby integrating power-law spectraconnecting the datapoints and extrapolating to the boundaries of common bands, and by

fittingthedatatovarioustemplatespectralenergydistributions derivedforAGNs.Table2includesmid-IRluminosities.These aregivenasthelogofthevalueinergs−1calculatedby power-law interpolation between the WISE bands, with a redward extrapolationtospanfrom3.4–42μmforcomparisonwiththe

IRAS-based FIR luminosity as listed in Keel et al. (2012a), which includes the range 42–122μm (Fullmer & Lons­ dale1989).SDSS1510+07isundetectedinthelongestWISE

band;thelistedluminosityassumesthetruefluxisatthislimit,

whilesettingthefluxtozerogivesaluminosity0.14smallerin the log. For comparison, we also list the L(FIR) value

(likewise, decimal log of the value in erg s−1) from Keel etal.(2012a).Manyoftheseobjectshavemid-IRluminosities muchgreater thantheFIR values,high enoughinMkn 1498 and the Teacup system to leave open the possibility of an obscured rather than fading AGN in the absence of other evidence (such as the recombination histories we calculate below).Intheothers,evenincludingtheMIRrangeoftenmost sensitiveto AGN heatingof surrounding grains, anorder-of­ magnitude energy mismatch is seen between the AGN itself and the requirement toionize distant clouds. ForUGC 7342 andSDSS 2201+11,theshortfallexceedsafactor100.These luminosity estimates are considered in comparison to ioniz­ ation-basedvaluesinSection 4.3.

2.4.SpectroscopicMapping

Weinvestigatethe kinematicsand ionizationstructure near the nuclei of several of these galaxies using integral-field opticalspectroscopyfromtheGeminiMultiple-ObjectSpectro­ meter(GMOS)system(Daviesetal.1997)atthe8mGillette Gemini-N telescope on Maunakea, under programs GN­ 2013B-Q-25 and GN-2014A-Q-25. The GMOS integral-field unit(IFU)samplestheskyin0 2 aperturescoveringaregion roughly3 5×5″inthesingle-slitmodewe used( Allington-Smith et al. 2002). The B600 grating covered a wavelength range usually encompassing from HeII λ4686 to the [SII]

(4)

Table3

GeminiGMOSIFUSpectra

Location Offsets (arcsec) UTdates FWHM (arcsec) Observers

UGC11185pos1 0.2W3.4S 2013Aug13,27 0.48 M.Hoenig

UGC11185pos2 2.5E1.2S 2014Aug13 0.41 A.-N.Chene

UGC7342pos1 1.7N 2014Jun26 0.79 J.Chavez

UGC7342pos2 1.7S 2014Jun24 0.76 M.Pohlen

Mkn1498 0 2014Jul26 0.40 L.Fuhrman

NGC5972 0.3E 2014Jun26 0.47 J.Chavez

TeacupENE 4.02E2.30N 2014Jul27 0.56 L.Fuhrman

TeacupE 2.78E0.04N 2014Jul18 0.44 M.Pohlen

TeacupSW 1.5W0.63S 2014Jun30 0.45 J.Ball

procedures described by Davies et al. (2015), including correctionforinternallyscatteredlightanduseofLACOSMIC

(van Dokkum 2001)to reject pixels contaminatedby cosmic rays.Thermsscatterofmeasuredcomparison-linewavelengths acrossthewholearraywas0.08Å,5 km s−1for[OIII],sothe accuracyofthiscalibrationismuchfinerthantheamplitudesof velocitystructures wetrace.Datafromeachobservationwere combined intoadatacubeevenlysampled onthesky, taking intoaccountatmosphericdispersion.Datacubesfromallthree grating positions were combined at the end to provide an additional levelof rejectionforcosmicraysandother sources ofbadpixels.FluxcalibrationusedasingleobservationofHZ 44;astheobservationsoccurredondatesfrom2013August13 to 2014 August 27, this flux calibration will be only approximate.Wedohaveexternal checksonlineratiosbased on previous long-slit spectroscopy (Keel etal. 2012a, 2015). Both UGC 11185 data runs were truncated by deteriorating conditions; we have only a single grating tilt for each IFU position, with 2×1200s exposures, and these are affected significantly bycosmeticissues, socoverageof somespectral linesisincompleteandtheirfluxscalesareunreliable.

Thedeliveredimagequalityrangedfrom0 4 to0 8FWHM duringtheseobservations,asestimatedfromred-light(r-band) acquisition images obtained immediately before the spectral data.MostoftheseobservationshadFWHM<0 5.Aslisted inTable3,wehaveobservationscenteredonthenucleiofMkn 1498andNGC5972,andsetsoftwooverlappingoffsetfields for UGC 7342 and 11185; the table includes the requested offsets from the galaxy nucleus to the observed IFU center point. FortheTeacupAGN, threeoffset fieldscovervirtually alltheemission-lineregionsseenintheHSTACSimages(Keel et al. 2015). Similar datafor this galaxytaken with the ESO VIMOS system have been shown by Harrison et al. (2015); theydescriberesultsonlyforthe[OIII]emissionlines,andthe spatial scale of 0 67 per pixel limits their spatial resolution compared tothese GMOSdata.

For straightforward comparison of emission-line properties overwiderangesinsignal-to-noiseratio(S/N), wefitGaussian profiles tothe lines, constraining Hα and the adjacent [NIII]

linestohavethesameFWHM,withthesameconstraintonthe

[SII]doublet.Insomeregions,eachlineisdoubleortriple,so we fit blended sets of Gaussians; we used the IRAF task splot interactivelyonthedatacubessampledin0 2 spatial increments. Line properties of interest (flux, wavelength, FWHM) were gridded into spatial maps for further analysis. Inmostcases,two-componentfitswereconstrainedtohavethe sameFWHM,forconsistencyacrossawiderangeinS/N;for thenarrowerprofilesdominatedbytheinstrumentalresolution, this is a good fit. Elsewhere, there may be systematics

introduced, but to first order the flux in each component is preserved.

3.IonizationGeometry

Some of these cloud systems show plausible ionization cones,brieflymentionedinPaperI.InFigure1weshow line-ratio images of [OIII]/Hα, which often map these structures moreclearly than[OIII]intensityalone.Amongourgalaxies, someshowcone-likestructuresinlineintensity;weseeafew wherethelineratiomapsionizationinaclearerstructure.Most notably,intheTeacupAGNthereisareasonablywell-defined triangular region of higher ionization (where the “hole” lies within it). In others, such as UGC 7342 and UGC 11185

(Hainline et al. 2016), the entire emission region could plausibly show a biconical structure without any internal differencesinionization.Inthesetwo systems,radial changes inthe line ratioare greaterthan azimuthalvariations at fixed projectedradius.

Broadly, all these emission complexes show a twofold symmetry,as sketchedinKeeletal. (2012a)fromearly data. Whentheirouterpartsresembleionizationcones,weoftensee much broader regions near the core (often completely encompassingtheAGN),whoseionizationmayhaveadistinct mechanism.

TheemissionregioninMkn1498ishighlyelongatedbutnot biconical. Amongthe remainder, mostof the opening angles implied by the HST data are close to our initial estimates, exceptfortheTeacupAGNwherewenowrecognizeaninner structuretracedbythe[OIII]/Hαratioforwhicheachhalfhas afullangleof37°.InSDSS2201+11,wederive25°,although thismaybeaffected byobscurationwithinthehostgalaxyon onesideofeachcloud.Therestarebroader,withNGC5252at 64°,UGC7342at75°,andNGC5972at96°toencompassthe outer structures.The broadestare seen inUGC 11185(116°) and SDSS 1510+07 (126°). Projection effects in a conical distribution make this angle appear larger than its three-dimensionalvalue.

Side-to-sideasymmetriesinionization(asinBPTplots,orin reconstructed ionization history) could in principle record front–backdifferencesinlighttraveltimewhenseenatasharp angle to the axis of ionization cones or radially elongated clouds. We do not actually see any such differences; our selection for large transverse extent favors cloud systems viewedneartheplaneofthesky.

(5)

Figure1.[O III]/Hαline-ratioimages,usinga“blackbody”colorpalettetoincreasethevisualdynamicrangeandtoshowfeaturesdimmerthaneasilyappearwitha puregrayscale.Thescaleisthesameforall,runningfromzerotoalineratioof3.6shownaswhite.Theimagesweremaskedat2σinHαintensity(theweakerlinein mostareas),thena3×3pixelmedianfilterwasusedtosuppressnoisesincethelineratioismoreconsistentacrosspixelsthantheintensities.Northisatthetopand easttotheleft;theangularscalevariesasshownby5″scalebars.ForthelargecloudsystemsofNGC5972andUGC7342,insetsshowwheretheenlargednuclear regionsarelocated.ThenuclearregionofSDSS2201isaffectedbyreddeningfromadustlane,sonoreliablelineratiosareavailableintheinnerparts.Thelocation ofeachAGNisshownbyaplussign.

Near the nuclei, and outside the ionization cones, several of these galaxies show gas with low ionization as seen in

[OIII]/Hα.OnegoaloftheGMOSIFUspectrawastomeasure additional lineratiosinthese regionstoprobetheirionization mechanism. Wewillexplore thisissueelsewhere,inconjunc­ tion with recentspectra from the HST Space Telescope Imaging Spectrograph that isolate key regions near selected nuclei.

Overall, the occurrence of ionization conespointing tothe nucleus suggeststhattheionizedgashasdistributionsthatare mostly radial, rather than, for example, ringlike. This means

thatprojecteddistance fromthenucleus canserveas auseful proxyforphysicalseparationfromtheAGN intheseclouds.

4.AGNLuminosityHistory

4.1. Method:RecombinationBalance

(6)

Figure 2. Radialplot of pixel-by-pixelsurface brightness (in instrumental units,detectedphotonspersecondperpixel) inHαfortheTeacupAGN,with superimposedcurvesofconstantQionin photonss−1.Median filteringover 3×3pixels was used to suppress residual cosmic-ray and charge-transfer effects, with the side effect ofcreating correlations between some sets of adjacentpoints.

Figure3.Radialplotofpixel-by-pixelminimumrequiredionizingluminosity fortheTeacupAGN,derivedfromthedatainFigure 2.Projecteddistancefrom the AGN isexpressed inlight-years to makethe timescalesapparent. The dashedlineatthetoprepresentstheionizationhistoryinferredindependently byGagneetal.(2014)fromdetailedmodelingofground-basedspectra.The envelopeatthebottomrepresentsa2σcuttosuppresstheeffectsofpixelnoise atlargeradii.

end-on,themeasuredluminosityinarecombinationlinegives alowerlimittotheionizingluminosityseenbythatvolumeof material.Thebrightestpixelsateachprojectedradiusgivethe most stringent limits. Even in these locations, the derived luminosity will bean underestimatebecauseof thecombined effectsofopticaldepthandcoveringfractionwithintheregion spannedbyeachpixelontheobject;bothquantities arelikely tohavecomplexandpatchydistributions.Suitablylocatedgas must be presenttoapply this approach, of course;astrongly

flaring AGN would look the same as one without suitable neutralgasaroundthehostsystem.

Despite these clear limitations, this approach could yield useful information on, for example, the history of AGN luminosity,iftheamountofunderestimationiscomparablefor the highest pixels atvarious radii(if notnecessarily between objects).RegionsfartherfromtheAGNarephysicallyunlikely

Figure4.InferredionizationhistoryofofPalomar–GreenQSOsfromBennert etal.(2002),calculatedasinFigure 7,showingpeakvaluesinbinsof1000lt­ yrwithradiallyadaptivesmoothing.Theprojectioncurveatthebottomshows theeffectofchangingthevectordirectionfromtheAGNtothegasby±30° fromtheplaneofthesky,forameasuredvalueatthebottomofthiscurve.

tointercept progressivelymore radiation, since the densityin emissionregionsgeneralfallswithdistancefromthecore,and the changes we will infer span orders of magnitude in the oppositesense towhatisexpectedfromthis. Twoimmediate checks on this approach are available—the photoionization modeling for spectra of various regions around the Teacup AGN(Gagneetal.2014),andnarrowband[OIII]structuresof aset ofPalomar–Green(PG)QSOs.

We employ the ionization balance as follows. We use the highest surface brightness observed in Hα as a function of projected distance from the AGN to estimate the required

(isotropic)emissionrateofionizingphotonsQionneededtopower the observed emission, making the broad assumption as noted abovethatthedensestgasclumpsaresimilaratvariousprojected distancesfromtheAGNwithineachsystem.Inpractice,weuse therequiredphotonratebetweentheHandHeionizationedges, sincethephotoionizationcrosssectionofheliumislargeenough to dominate the absorption immediately above 54.6eV. As expected,theHSTimagesrevealnumeroussmallregionsofhigh surfacebrightness,sothatthelowerlimitwederiveinthiswayis often much higher than estimated from ground-based spectra

(Keel et al.2012a).Our calculation incorporatesthe fractionof recombinations leading to an Hα photon (Spitzer & Green-stein1951;Nussbaumer&Schmutz1984). In“CaseB,”where thegas isoptically thick toLyα,thefractionofrecombinations leading to an Hα photon ranges roughly from 0.25 to 0.34 dependingonthetrappingpropertiesofLyα.Theotherlimit,case Awherethenebula isopticallythininLyα,has thesefractions lower by a factor 1.6 (these values are for Te=104K, with recombinationcoefficientsascollectedbyOsterbrock&Ferland 2006).Ifwedenotethefractionofrecombinationsleadingtoan Hαphotonbyf,Qioncanbederivedconvenientlyfromthe area-normalized count rate N. This value represents the arrival rate

(7)
(8)

Figure 6.Inferredionization historyofextendedAGN-ionizedcloudsasin Figure 3,withlinesshowingthepeakvaluesinevery1000lt-yrbinforeach object.Theprojectioncurveatthebottomshowstheeffectofchangingthe vectordirectionfromtheAGNtothegasby±30°fromtheplaneofthesky. The largepoints near zero radius show the derived observed-epoch AGN luminosityfromHαatthenuclei,exceptforIC2497,whichfallswelloffthe scaleat1.1×1050photonss–1.

unobscuredtelescopeareaA,sothenumberemittedpersecondin theemitter’s frameis4pD2(1 +z)NA h forsourcedistanceD,

now includingthe time dilation (1+z) in photon arrivalrate. Neglecting projectioneffects,whichare poorly known,material ineachpixelofobservedsolidangleawouldinterceptafraction of theemittedradiation of theorder ofa/4πR2where Ris the projecteddistancefromthenucleustotheareaofinterest;ifRis inpixels,a=1forunsmootheddata.

Collectingthesecomponents, thecountrateNperunitarea inasingle pixelimplies

2 2

Qion =(1 +z)(N hf)4pD 4pR

foraregionprojectedRpixelsfromtheAGN.Fortherangesof densityandLyαopticaldepththatmatterhere,asnotedabove,f

ranges from 0.16 to0.34. This value of Qionis alower limit, sincewethenhavenoconstraintonhowmuchradiationescapes the system. To be conservative, we discuss values using f =

0.29,inthemiddleoftherangeofCaseBphotoionization.This was cast in asomewhat different way in Keelet al. (2012a), usingtheangularwidthofaspectrographslitaboutthenucleus atagivenR.

TheACS ramp-filter passbandhasanearlyflatprofilenear the peakwavelength,falling below50% ofpeaktransmission welloutsidethespanofthe[NIII]lineswhencenteredonHα attheseredshifts.Wemakeacorrectionfortheadjacent[NIII]

emission based on our slit spectroscopy (Keel et al. 2012a), assuming the behavior of the line ratio to be circularly symmetricaboutthenucleioneachside.Wefitthiscorrection factorasapolynomialinradiusforcomputationalconvenience. For optically thin gas, or material with a small covering factor (when shadowing byabsorberscloser tothenucleus is notimportant),therewillbeanupperenvelopeinthediagram ofHαsurfacebrightness-projectedradius,whichwillfollowa 1/r2 form for constant source luminosity as long as various regions inemissiondo nothave radically differentprojection factors.Departuresfromthisbehaviorcanbeevidenceoflong­ termchangesinQion;neglectingprojectionfactors,onecanin principle extractthesourcehistory fromsuchadiagram.

Figure7.InferredionizationhistoryofextendedAGN-ionizedcloudsshowing thepeak valuesinevery1000lt-yrbinforeachobject.Thisrepresentation smoothsthederivedQvaluesbyradialamountsintendedtoreflectchangesin the recombination timescale withradius, tobetter distinguishstructure that couldbeduetoAGNvariationfromgasstructure.Theprojectioncurveatthe bottomshowstheeffectonapointatthebottomofthecurve,ofchangingthe vectordirectionfromtheAGNtothegasby±30°fromtheplaneofthesky. The large points near zero radius show the derived observed-epoch AGN luminosityfromHαatthenuclei,exceptforIC2497,whichfallswelloffthe scaleat1.1×1050photonss–1.

4.2.ConsistencyTests

Weshow onesetof dataforthis calculationsuitableforan externaltestinFigure2fortheTeacupAGN,wherethesurface brightness is in electrons pixel−1 s−1 and the radius is in arcseconds.Transformingtheseintothephysicallymeaningful units of Qion and projected distance in light-years gives our estimatesoftheminimumQion ateachpixel.

This conversion is illustrated in Figure 3, showing Qion converted into ionizingluminosity for each pixelabove a 3σ thresholdfortheTeacupAGN.Theupperenvelopeisthemost significantfeature,sincewe arguethatthe densestregionsare likelytobecomparableatvariousradii.Forthisobject,wecan compare our results to the detailedphotoionization modeling usingthe CLOUDY90code(Ferlandetal. 1998) carriedout byGagneetal.(2014)usingground-basedspectraspanningits emissionregions. Thoseresults paralleltheupperenvelopeof ourreconstructed points,systematically≈25%higherdepend­ ingon howthe envelopeisdefined, which supports both our resultsandthebasic assumptionofouranalysis,thatthemost opticallythickregionsarecomparableatallradiiintheclouds

(so we tentatively assume this to be the case for the other galaxies inoursample).Inparticular, ouranalysisreproduces thesamechangeinluminosity,decliningbyafactor101.03over a time ΔT=26,500yr between the peaks in reconstructed luminosity.Itisalsoreassuringthatthepeaksinreconstructed luminosity approach the spectroscopic results closely but do notexceedthematanypoint.

(9)

Table4

TimescaleEstimatesforVariabilityEpisodes

Galaxy DataSpan(yr) TimeRange(yr) Behavior e-foldingTimescale(yr)

IC2497 65,000 111,000–102,000 brighten1.0dex 4000

102,000–60,000 constant L

60,000–46,000 fade2.0dex 3000

Mkn1498 32,000 32,000–14,000 brighten0.6dex 13,000

14,000–2000 fade0.5dex 10,000

2000–0 fade1.6dex 550

NGC5252 40,000 40,000–14,000 constant L

14,000–2000 fade1.2dex 4300

2000–0 fade1.4dex 620

NGC5972 55,000 55,000–15,000 fade0.3dex 5800

15,000–5000 fade1.0dex 4300

5000–3000 brighten0.5dex 1700

3000–0 fade1.3dex 1000

SDSS1510 60,000 60,000–18,000 constant L

18,000–2000 fade1.8dex 3900

2000–0 fade0.6dex 1500

SDSS2201 45,000 45,000–10,000 fade1.2dex 13,000

10,000–3000 fade0.8dex 3800

3000–0 fade0.6dex 2200

Teacup 55,000 55,000–36,000 brighten1.0dex 8000

36,000–1000 fade1.2dex 12,000

(Gagneetal.) 39,000 39,000–0 fade1.7dex 9800

UGC7342 120,000 120,000–50,000 constant L

50,000–34,000 fade1.03dex 6700

34,000–20,000 fade0.1dex 60,000

20,000–11,000 fade0.40dex 9700

11,000–2000 fade0.6dex 6500

2000–0 fade>1.0dex 900

UGC11185 55,000 55,000–45,000 brighten1.0dex 4300

45,000–4000 constant L

4000–0 fade2.2dex 800

Table5

NuclearHαFluxesandIonizingRates

Galaxy Hαflux Qion MIR+FIRQion

(ergcm−2s−1) (photonss−1) (photonss−1)

IC2497 1.3×10−15 2.6×1050 2.9×1054

Mkn1498 5.5×10−14 3.3×1052 9.3×1054

NGC5252 4.5×10−13 8.4×1052 8.3×1053

NGC5972 2.6×10−14 8.4×1052 8.1×1053

SDSS1510 5.9×10−14 4.4×1052 <1.4×1054 SDSS2201 1.1×10−13 3.2×1052 <2.2×1053

Teacup 7.9×10−14 2.0×1053 9.5×1054

UGC7342 1.3×10−13 1.0×1053 1.8×1053

UGC11185 4.9×10−14 2.9×1052 1.2×1054

the WFPC2 linear-ramp filters and continuum subtraction incorporating stellar point-spread functions. The scaling of their implied ionizing luminosities is only approximate; we used a mean value for [OIII]/Hα=4, lacking independent narrow-lineratiosatthecore.Thatprogramwastargetedatthe sizes of narrow-line regions, with no selection for extent of EELRs,andonlythreeofthePGQSOshavedetectedemission beyondthe10kpcradialboundusedtoselectourfading-AGN sample. We masked regions around obvious companion galaxies, but not a bright emission region within 1″ of PG 0157+001sinceitsnatureisambiguous.Theimpliedhistories are shown in Figure4. These show brightening, fading,and near-constant episodes. PG 0157+001 and PG 1012+008

show fading by nearly an order of magnitude before bright­ ening close to the observed epoch, while PG 0953+414 has brightened bynearly this much.Brighteningepisodes willbe harder todetect atlarge radii because the signal drops more strongly near the detection threshold. This test broadly confirms that our sample of fading candidates does indeed show distinct behavior using our techniqueof reconstructing luminosityfromrecombination balance.In particular,noneof thePGQSOs showsthe rapiddropsinthelast 20,000yrthat areubiquitousinourfading-AGNsample.

4.3. LuminosityHistoriesofFadingAGNs

Thetwoexternaltestsoutlinedabovesuggestthat,indeed,we canreconstructatleastqualitativehistoriesforAGNluminosity using pixel-by-pixel recombination balance. Figure 5 shows these pixel-by-pixel results for the other seven objects in our sampleplusIC2497andHanny’sVoorwerp,wheresimilardata were describedby Keeletal. (2012b).Figure 6 compares the upperenvelopesforallobjects,connectingthebrightestpixelsin binsof1000lt-yr.

(10)

Figure8.Continuum-subtractedHα+[N III]imagesofnuclearregions,mostshowingloopsorringlikeemissionstructuressuggestingoutflow.InNGC5972and SDSS2201,colorchangesfromdustlaneslimittheprecisionofcontinuumsubtractionwiththeavailablefilters.Northisatthetop,easttotheleft;scalebarsare1″ long.Thethreeobjectsinthebottomrowhaveverybrightcentralemission,sotheirimagesareshownwithlogarithmicintensityscales;therestuselinearscales.

roughlyaccountsforadecreaseindensitywithradiusmatching the limitsfromLintott etal. (2009).The pointsatΔT=0 in bothversionsofthisfigurearefromtheHαfluxesatthenuclei

(operationally within a projected radius of 1″ from slit spectroscopy), in the approximation that they represent full covering of the instantaneous ionizing flux (strictly, a lower limitsinceweseethationizingphotonsdoescape thenuclear region). The flux values are taken from our spectroscopy in Keel et al. (2015), as given in Table 5, including only the narrow component of the composite profile in Mkn 1498 for consistency. As a rough numerical description of these reconstructedhistories,wecollectinTable4asetofproperties

fit exponentially to segments of the smoothed version in Figure 7, showing not only fading periods but constant and brightening episodes.

Weconsideredseveralalternativeapproachestocalculating ionizing history, including integrating the line flux in slices across the putative ionization cones. This was defeated by

the very patchy structure of the ionized features; tracing only the peak intensity reduces sensitivity to the larger-scale structure.

(11)
(12)

Figure9. (Continued.)

limits for most of these AGNs (listed in Table 5) fall well fromvariabilityratherthanunusuallyhardFUVspectralslopes below even the lower limits imposed by recombination orlocalizedextinctionaroundtheAGN.TheshortfallsforQion balance, adding to the case for drops in luminosity. Finally, derived fromthe infrared are less extreme, indicating modest the spatially resolved drops in required ionizing flux seen localobscuration,butinmostcasesstillverylarge.

(13)

Figure9. (Continued.)

(14)

Figure9.(Continued.)

5.AGNOutflowsandFeedback

In IC 2497, host galaxy of Hanny’s Voorwerp, HST

observations reveal an expanding loop of low-ionization gas extending ≈500pc to one side of the nucleus (Keel et al. 2012b).This, along with evidencefor an oppositelydirected radio jet (Józsa et al. 2009; Rampadarath et al. 2010) and bubble-like structure in the X-ray gas (Sartori et al. 2016), invites speculation that these objects are not necessarily undergoing a near-shutdown of accretion, but may (also) be switchingmodesofenergyoutputincidentaltotheaccretion,to become kinematicallydominated(sometimesknownas radio-mode).In X-ray binaries, the switchfromradiative tokinetic modeisassociatedwithachangeinaccretionstate(andthusa decrease in the luminosity). It is possible that the same is happening also in AGNs (Sartori et al. 2016). There are additionalAGNswithwell-studiedbubblesor loopsattributed tooutflows,such as recentresults onNGC 3393byMaksym et al. (2017). This motivated us to look for similar loops of emission, and seek kinematic evidence as to whether they representoutflowsorsomethingelse.

Morphologically similar loops to one side of nuclei, or encircling them, are common in this sample (Figure 8), and could by themselves be signatures of outflow episodes. However, the GMOS IFU datagive amore complex picture both kinematicallyand inemission-line ratios.

Kinematic signs of outflow do appear in several nuclei, although some of the loops are more nearly in rotation than radial motion. We can evaluate this using the Gemini IFU spectroscopy, as well as the wider-field Fabry–Perot velocity

fields for [OIII] obtained with the 6m BTA (Paper I). Spectroscopic signatures of outflow include multiple-peaked or asymmetric line profiles, and departures from large-scale rotationalvelocityfieldsevenforlocallynarrowandsymmetric

profiles.Suchfeaturesareseeninconfinedregionsofsomeof ourtargetgalaxies.Figure9overlaysthe[OIII]profilesforthe Gemini IFU dataon HST narrowband images in[OIII],with the line profiles representing averages of the data over 0.5×0 5 regions.

In Mkn 1498 the ringlike emission features that are prominent at Hα are dominated kinematically by rotation, possibly with one outflowing feature adjacent to the nucleus

(Figure10).Theimpliedrotationaxisisclosetotheminoraxis ofthelargestemission-linestructures,makingitalmostparallel tothelarge-scaleradiostructure.Inthelineprofiles,redwings occuralongtheouterringtothesouth,andabout2″eastofthe nucleusalongthe otherring(butnottherest ofit).Theinner arc structure to the northwest of the nucleus hasnarrow line profiles,butdepartsfromthekinematicpatternofthevelocity peakselsewhereinthering.

As seen in Figure 9, UGC 7342 shows substantial regions withdoubleemission-linepeaks,likelyevidenceofoutflowinat leastonecomponent.Weexaminethisinmoredetailbyfitting setsoftwoGaussian[OIII]profileswherethisimprovesthefit, constrainingthemtohaveequalwidthsforconsistencyacrossa rangeofS/Ns.SimilarbehaviorisseeninHαandHβ, butHβis weakerandHαisaffectedbyoverlapwith[NIII]components. The results are shown in Figure 11. Only peak separations

(15)

Figure10.SummaryofGMOSIFUresultsintheinner3.4×4.8arcsecofMkn1498.Northisontheright,eastatthetop.TheHSTnarrowbandimageincludesboth Hαand[N III]lines.Forallquantities,lightershadesrepresentlargervalues.

(16)

Figure11.Resultsoftwo-componentGaussian-blendfitsto[O III]emissioninUGC7342.TheHSTACS[O III]imageisscaledlogarithmicallytomorefullydisplay thestructuresnearthecore.ThetwoGMOSIFUfieldsareabuttedattheirnominallocationshere,givingcontiguouscoverageover4 8×6 4.Northisatthetop, easttotheleft.Theintensitiesofredandblue(showningreenforcontrast)componentsarescaledindependentlyforbettervisibility.SomeregionsoflargeFWHM areadjacenttodouble-peakedareas,presumablyrepresentingunresolvedmultiplecomponents.Forcomparison,theleftpanelshowstheFabry–Perotvelocity fieldof theentiregalaxyasmeasuredwiththeBTA(Keeletal. 2015),combininghueforradialvelocitywithluminancefor[O III]flux.

theBTAdatajusttothenorthofthenucleusareagoodmatch toregionsoflinesplittingintheGMOSdatacubes.

Weareleftwithacomplexempirical pictureforkinematics in the inner fewkiloparsecs of UGC 7342. Emission-line componentssuggestoutflow,oratleastmotionsdepartingfrom the overall rotation pattern by as much as 260kms−1, in regions of diffuse, low-surface-brightness emission perpend­ icular to the main emission regions, with the strongest off-rotational emissionintheouter partsofabrightcloud,partof whichalsoshowsbluewingsbeyondasimpletwo-component lineprofile.Wemightspeculatethatthiscouldbeentrainment ofdensergasbyalow-densityoutflowsuchasseenwestofthe nucleus.Curiously,thebrightest[OIII]structure,resemblingan inclined ring just southwest of the AGN position, has no distinct signatureinradial velocity.

The Teacup AGN was shown to have resolved outflow signatures by Harrison et al. (2015). We find double or asymmetricprofiles extending 2″north, 4″west,>1 5 south, and 2″ east in a roughly circular zone. What appears as an

[OIII] peak in the center of the loop on the underlying HST

image may be affected by residuals from cosmic-ray events. TheGeminidataarebetteratdetectingemissionthatisdiffuse and atlowsurfacebrightness,asinthecenteroftheloop,and show no distinct emission-line component there. Blue wings dominate eastof thenucleus, redwings tothe west;inmany

locations they extend to 1000kms−1 from the systemic velocity.

NGC5972showssignificantvelocitystructure,andanearly constant ionization level (Figure 12). It shows double and asymmetricpeaksnear thedustlane,which canoccurstrictly fromobscurationinacontinuousvelocityfield,andintheinner partoftheemission-linelooptoitseast.Overall,the emission-line loop to the east of the nucleus shows no discernible velocitysignature;it does havenarrower lineprofiles thanits surroundings,perhapsbecauseitsstrongeremissiondominates themorediffusegaselsewherealongthelineofsight.Thereis larger-scale velocity structure in the GMOS field spanning

≈100kms−1,not matchingthe loop morphologyor location; thisispartofthedeparturesfromafittocircularmotionsseen nearthenucleusinthewide-fieldBTAFabry–Perotmap(Keel etal.2015).The[OIII]/Hαratiodeclineswithradiusfromthe nucleus atthesame rateintheloop and itssurroundings;the loopdoesnotstandoutinthisparameter. Ithaslower[NIII]/

Hαand[SII]/Hα ratiosthanitssurroundings,butnot[OIII]/

Hαorthedensity-sensitive[SII]λ6717/λ6731ratio.Theloop region 2 0 eastof the nucleus has [NIII] λ6583/Hα =0.33

(17)

Figure12.SummaryofGMOSIFUresultsintheinner3 4×4 8ofNGC5972.Northisatthetop,easttotheleft.TheHSTnarrowbandimageincludesbothHα and[N III]lines.ThepeakFWHMof326kms−1occurs0 2eastofthepeakHαflux.Forcomparison,theleftpanelshowstheFabry–Perotvelocityfieldoftheentire galaxyasmeasuredwiththeBTA(Keeletal. 2015),combininghueforradialvelocitywithluminancefor[O III]flux.

1998).Verybroadly,thelineratiosshowlowerabundancesin this small-scale loop of emission than in its surroundings, whichcouldsuggest,contrarytoitsappearance,thatitconsists ofinfallingorrecentlyaccretedgas.Together,theseproperties seem more like infall than an AGN-driven outflow; the alternative would require a fairly contrived set of circum­ stances,suchasmotionintheplaneoftheskyandtheoutflow itself beingunseen(forexample,toohotforopticalemission) and entraining material that originated in a low-luminosity companion disrupted by the obvious merger history. Young stars might give the observed line ratios, in which case it would be acoincidence thatthe [OIII]/Hα ratio matches the surroundingAGN-ionizedgassoclosely,butnoluminousstar clustersappearinWFC3continuumimages(Keeletal.2015).

6.Comparison:Changes inLuminosity inAccretion

Simulations

We cancomparethe changesin luminosity we measureto simulations such as Novak etal. (2011) tosee howcommon suchdramaticchangesarepredictedtobeforvariousfeedback properties. Existing simulations have time resolution of the orderof104yr,sotheremaybejustenoughinformationtotell

what behavior is expected on timescales close to 105 yr. Figure 13 shows how common various levels of drops in accretionluminosity (derived fromthesimulationsas L/LEdd) are predictedto be when sampled on varioustimescales. We emulatethemeasurementsbysmoothingthestartingluminos­ ityateach time stepby30% of the timespan beingsampled, whichmimicstheeffectsoffinitedepthalongthelineofsight and increased recombination timescale at the typically lower densitiesfartherfromtheAGN.Althoughthetimeresolutionis notreallyadequatetotestwhathappenswhensamplingafew 104yr,itisencouraging that thissimulation showsafew per centofitstimestepsthatwouldsatisfyourselectioncriteriafor AGNfadingifsuitablesurroundinggaswerepresent(Table6). Thetableliststhefractionoftime stepswheretheluminosity drops at least as steeply as the listed ratios, evaluated for various valuesof ΔT according to L( )T L(T- DT) where theaveraging inthe denominatorisover0.3ΔT asabove. As Novaketal. (2011)note,theirsimulations showlong periods of near-constant accretion, and other periods showing many cyclesofnearlyperiodicchanges.

(18)

Figure13.FractionoftimestepsfromsimulationA05inNovaketal.(2011)

as a function of the drop in luminosity L( )T L(T- DT), evaluated on timescales from 5×105 to 5×106 yr. For each time series,the starting luminositywassmoothedover30%oftherelevanttimespantoroughlymodel theeffectsoffiniteclouddepthalongthelineofsightandlongerrecombination timesatlowdensities.Thelackofpointstothelowerrightindicateseffectsof timeresolution;at longertimescaleswemaybeseeingeffectsofphysically driventimescalesinthesimulation.

due to feedback from the AGN) yield faster changes in accretionrate(Gabor&Bournaud2013).

7. Summary

WepresentHST narrowbandimagingandGemini

integral-field optical spectra for a set of fading AGNs, selected for shortfalls in the energy budget between giant ionized clouds and the nuclei themselves, suggesting significantreductionin ionizingphotonswithinthelast≈2×104yr.Inparticular,we use recombinationbalance toestimatethehistory ofradiative output, findingthat thefeatureincommon isaradial dropin luminosity within 20,000 yr before our direct view of the nucleus. For earlier times, we see a range of behaviors— constant, brightening, fading—which are echoed in similarly reconstructedhistoriesofPGQSOswhere[OIII]HSTimaging isavailable.

In contrasttomanyof theextendedemissioncloudsaround radio-loud AGNs, these are rotationally dominated and show onlyverylocalizedoutflowvelocitiesgreaterthan≈100kms−1. These EELRs are mostly externally illuminated tidal debris ratherthanwindmaterial(Keeletal.2015).

As inKeel etal. (2012a),these results continuetosupport theideathatAGNswithextendedemissionregionsarebright for periodsof 104–105 yrat atime,with substantially fainter episodesinterspersed(whichfitswiththeideathattheopposite behavior—brightening—canbeseen inthe number of X-ray­ bright AGNs without substantial narrow-line regions; Scha­ winski etal.2015).Mostgalaxiesdonothavelargereservoirs of (extraplanar) cold gas to show such behavior, so we can trace itpreferentially ininteractingormerging systemswhere warpeddiskgasortidaltailsprovideascreentobeionizedby escapingionizingphotons.

Among EELR hosts from the Galaxy Zoo sample, fading AGNs make up ≈40% of the total, leading to the simple estimate of long accretion timescales in Keel et al. (2012a). This leaves open the possibilitythat fading cases could be a specificsubsetof thewholepopulation. Forexample,inspiral

Table6

SimulatedTimeFractionsinLowStates

Log(Luminos­ ityDrop)

Timespan 5×106yr

Timespan 2×106yr

Timespan 106yr

Timespan 5×105yr

−1 0.26 0.20 0.16 0.16

−2 0.15 0.12 0.085 0.063

−3 0.098 0.067 0.050 L

−4 0.063 0.037 0.015 L

−5 0.039 0.019 0.0005 L

Note.Fractionoftimestepsinthesimulationwherethemeasuredluminosity dropwithrespecttotheaveragedpastvalueismoreextremethanthereference valueforeachcolumn.

ofboundbinarysupermassiveblackholescoulddisturbapre­ existing accretion disk, while significant changes in the directionofionizingradiation couldalsoresultfromthekinds of tilted disk discussed by Lawrence &Elvis (2010). In that case, shadowing by the disk itself would leave ionization “cone”shapesthatareset bytheintersectionof twoconesof differentcenterdirections andopeningangles.

This work was supported by NASA through STScI grants HST-GO-12525.01-Aand-B.Someofthedatapresentedinthis paper were obtained from the Mikulski Archive for Space Telescopes (MAST). This research has made use of NASA’s Astrophysics Data System,and the NASA/IPAC Extragalactic Database (NED), which is operated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics and Space Administration. We thank Linda Dressel for advice on setting up the HST

observations,especiallyreducingintrusivereflections.Identifica­ tion of thisgalaxy sample was possible through theefforts of nearly 200 Galaxy Zoo volunteers; we are grateful for their contributions,andthankoncemorethelistofparticipantsinKeel etal.(2012a).Thisworkwaspartlysupportedbyagrantfromthe President of the Russian Federation (MD-3623.2015.2). The observationsobtainedwiththe6 mtelescopeoftheSAOofthe RASwerecarriedoutwiththefinancialsupportoftheMinistryof Educationand Science oftheRussianFederation(contractsno. 16.518.11.7073 and 14.518.11.7070). We benefited from con­ versations with Hai Fu, Larry Rudnick, Kelly Holley-Bock­ elmann,TamaraBogdanovic,andStephanieJuneau.GregNovak kindlyprovideddetailedresultsofhisnumericalsimulations.We thankDaraNormanandKathyRothforkeyhelpinsettingupthe Gemini observations, James Turner for sharing experience in subtleties of their processing, and Gemini staff observers M. Hoenig,A.-N.Chene,J.Chavez,M.Pohlen,L.Fuhrman,andJ. Ballforobtainingthesedatainqueuemode.

(19)

ZoowasmadepossiblebyfundingfromtheJimGrayResearch FundfromMicrosoftandTheLeverhulme Trust.

Thispublicationmakesuseofdataproductsfromthe

Wide-fieldInfrared Survey Explorer, whichis ajoint projectof the University of California, Los Angeles,and the Jet Propulsion Laboratory/CaliforniaInstitute of Technology,funded by the NationalAeronautics andSpace Administration.

Facilities: HST (ACS, WFC3, WFPC2), BTA, Gemini Gillette, WISE.

References

Allington-Smith,J.,Murray,G.,Content,R.,etal.2002,PASP, 114,892

Bennert,N.,Falcke,H.,Schulz,H.,Wilson,A.S.,&Wills,B.J.2002,ApJL,

574,L105

Bland-Hawthorn,J.,Maloney,P.R.,Sutherland,R.S.,&Madsen,G.J.2013,

ApJ, 778,58

Cutri,R.M.,Wright,E.L.,Conrow,T.,etal.2014,ExplanatorySupplementto the WISE, Section 4.4, All-Sky Data Release Products, http://wise2. ipac.caltech.edu/docs/release/allsky/expsup/sec4_4h.html

Davies, R.L., Allington-Smith, J.R., Bettess,P.,et al. 1997, Proc.SPIE,

2871,1099

Davies,R.L.,Schirmer,M.,&Turner,J.E.H.2015,MNRAS, 449,1731

Ferland,G.J.,Korista,K.T.,Verner,D.A.,etal.1998,PASP, 110,761

Fullmer,L.,&Lonsdale,C.J.1989,CatalogedGalaxiesandQuasarsObserved intheIRASSurvey, JPLD-1932

Gabor,J.M.,&Bournaud,F.2013,MNRAS, 434,606

Gagne,J.P.,Crenshaw,D.M.,Kraemer,S.B.,etal.2014,ApJ, 792,72

Guo,F.,&Mathews,W.G.2012,ApJ, 756,181

Hainline,K.N.,Hickox,R.C.,Chen,C.-T.,etal.2016,ApJ, 823,42

Harrison,C.M.,Thomson,A.P.,Alexander,D.M.,etal.2015,ApJ, 800,45

Husemann, B., Bennert, V. N., Scharwächter, J., Woo, J.-H., & Choudhury,O.S.2016,MNRAS, 455,1905

Józsa,G.I.G.,Garrett,M.A.,Oosterloo,T.A.,etal.2009,A&A, 500,L33

Keel,W.C.,Chojnowski,S.D.,Bennert,V.N.,etal.2012a,MNRAS, 420,878

Keel,W.C.,Lintott,C.J.,Schawinski,K.,etal.2012b,AJ, 144,66

Keel,W.C.,Maksym,W.P.,Bennert,V.N.,etal.2015,AJ, 149,155

Lawrence,A.,&Elvis,M.2010,ApJ, 714,561

Lintott,C.J.,Schawinski,K.,Keel,W.,etal.2009,MNRAS, 399,129

Lintott,C.J.,Schawinski,K.,Slosar,A.,etal.2008,MNRAS, 389,1179

Liu,G.,Zakamska,N.L.,Greene,J.E.,Nesvadba,N.P.H.,&Liu,X.2013,

MNRAS, 430,2327

MacLeod,C.L.,Ross,N.P.,Lawrence,A.,et al.2016,MNRAS,inpress

(arXiv:1509.08393)

Maksym, W. P., Fabbiano, G., Elvis, M., et al. 2017, ApJ, in press

(arXiv:1611.05880)

Menezes,R.B.,Steiner,J.E.,&daSilva,P.2016,ApJ, 817,150

Morse,J.A.,Cecil,G.,Wilson,A.S.,&Tsvetanov,Z.I.1998,ApJ, 505,159

Muno,M.P.,Baganoff,F.K.,Brandt,W.N.,Park,S.,&Morris,M.R.2007,

ApJL, 656,L69

Netzer,H.1990,ActiveGalacticNuclei(Heidelberg:ARI), 57

Novak,G.S.,Ostriker,J.P.,&Ciotti,L.2011,ApJ, 737,26

Nussbaumer,H.,&Schmutz,W.1984,A&A, 138,495

Osterbrock, D.E., & Ferland, G. J. (ed.) 2006, Astrophysics of Gaseous NebulaeandActiveGalacticNuclei(2nded.;Sausalito,CA:Univ.Science Books), 22

Penston,M.V.,&Perez,E.1984,MNRAS, 211,33

Prieto,M.A.,&Freudling,W.1996,MNRAS, 279,63

Rampadarath,H.,Garrett,M.A.,Józsa,G.I.G.,etal.2010,A&A, 517,L8

Röttgering, H.J. A.,Tang, Y.,Bremer, M. A.R., et al. 1996, MNRAS,

282,1033

Ruan,J.J.,Anderson,S.F.,Cales,S.L.,etal.2015,arXiv:1509.03634

Runco,J.N.,Cosens,M.,Bennert,V.N.,etal.2016,ApJ, 821,33

Runnoe,J.C.,Cales,S.,Ruan,J.J.,etal.2016,MNRAS, 455,1691

Sartori,L.F.,Schawinski,K.,Koss,M.,etal.2016,MNRAS, 457,3629

Schawinski, K., Koss, M., Berney, S., & Sartori, L. F. 2015, MNRAS,

451,2517

Schirmer,M.,Diaz,R.,Holhjem,K.,Levenson,N.A.,&Winge,C.2013,ApJ,

763,60

Schirmer, M., Malhotra, S., Levenson, N.A., et al. 2016, MNRAS, 463, 1554

Schweizer,F.,Seitzer,P.,Kelson,D.D.,Villanueva,E.V.,&Walth,G.L. 2013,ApJ, 773,148

Shankar,F.,Weinberg,D.H.,&Miralda-Escudé,J.2009,ApJ, 690,20

Spitzer,L.,Jr.,&Greenstein,J.L.1951,ApJ, 114,407

Stockton,A.,Fu,H.,&Canalizo,G.2006,NewAR, 50,694

Storchi-Bergmann,T.,Schmitt,H.R.,Calzetti,D.,&Kinney,A.L.1998,AJ,

115,909

vanDokkum,P.G.2001,PASP, 113,1420

Véron-Cetty,M.P.,&Véron,P.2001,A&A, 375,791

Wright,E.L.,Eisenhardt,P.R.M.,Mainzer,A.K.,etal.2010,AJ, 140,1868

Figure

Updating...

Related subjects :