An Active Galactic Nucleus Caught in the Act of Turning Off and On
Julia M. Comerford, R. Scott Barrows, Francisco Müller-Sánchez, Rebecca Nevin, Jenny E. Greene, David Pooley, Daniel Stern, Fiona A. Harrison
AApJ accepted
Preprint typeset using L A TEX style emulateapj v. 03/07/07
AN ACTIVE GALACTIC NUCLEUS CAUGHT IN THE ACT OF TURNING OFF AND ON
Julia M. Comerford , R. Scott Barrows , Francisco M¨uller-S´anchez , Rebecca Nevin , Jenny E. Greene ,David Pooley , Daniel Stern , and Fiona A. Harrison Department of Astrophysical and Planetary Sciences, University of Colorado, Boulder, CO 80309, USA Department of Astrophysical Sciences, Princeton University, Princeton, NJ 08544, USA Department of Physics and Astronomy, Trinity University, San Antonio, TX 78212, USA Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Drive, Pasadena, CA 91109, USA and California Institute of Technology, 1200 East California Boulevard, Pasadena, CA 91125, USA
ApJ accepted
ABSTRACTWe present the discovery of an active galactic nucleus (AGN) that is turning off and then on againin the z = 0 .
06 galaxy SDSS J1354+1327. This episodic nuclear activity is the result of discreteaccretion events, which could have been triggered by a past interaction with the companion galaxythat is currently located 12.5 kpc away. We originally targeted SDSS J1354+1327 because its SloanDigital Sky Survey spectrum has narrow AGN emission lines that exhibit a velocity offset of 69 kms − relative to systemic. To determine the nature of the galaxy and its velocity-offset emission lines,we observed SDSS J1354+1327 with Chandra /ACIS,
Hubble Space Telescope /Wide Field Camera 3,Apache Point Observatory optical longslit spectroscopy, and Keck/OSIRIS integral-field spectroscopy.We find a ∼
10 kpc cone of photoionized gas south of the galaxy center and a ∼ < yrs later the second AGN outburstlaunched the northern shock front. SDSS J1354+1327 is the galaxy with the strongest evidence foran AGN that has turned off and then on again, and it fits into the broader context of AGN flickeringthat includes observations of AGN light echoes. Subject headings: galaxies: active – galaxies: nuclei INTRODUCTIONWhen a supermassive black hole (SMBH) accretes asufficient amount of gas, it shines as an active galacticnucleus (AGN). However, this is not a continuous pro-cess; SMBHs can turn on and off as AGNs depending ontheir accretion histories. A local example is the SMBHin the Milky Way, which is currently quiescent, but dif-fuse bubbles of gamma-ray emission extending ∼
10 kpcabove and below the Galactic plane indicate that theSMBH was active < yrs ago (Su et al. 2010). Evi-dence of this past activity is also seen in X-ray reflectionoff galactic center molecular clouds (e.g., Koyama et al.1996; Zhang et al. 2015).Typical AGN variability timescales have been inferredto be 10 to 10 yrs between turning on and turning off(e.g., Schawinski et al. 2015). This time variability islinked to accretion of clumpy interstellar medium (ISM)in individual events, as seen in simulations of SMBH ac-cretion and the corresponding energetic AGN outbursts(e.g., Hopkins & Quataert 2010; Gabor & Bournaud2013, 2014). Since observable emission signatures canlinger for up to 10 yrs after an AGN has turned off ina galaxy, this leads to observable light echoes from pastAGN activity, which are seen in systems such as voor-werpen and changing-look AGNs (e.g., Tremonti et al.2007; Lintott et al. 2009; F¨orster Schreiber et al. 2014;Keel et al. 2015; LaMassa et al. 2015; Runnoe et al. 2016;Schirmer et al. 2016).Here we present a galaxy that has evidence of two dis-crete AGN outburst events. SDSS J135429.05+132757.2(hereafter SDSS J1354+1327, located at z = 0 .
06) wasoriginally noted for its velocity-offset narrow emission lines. By analogy to double-peaked narrow emission linesused to identify dual AGN candidates (e.g., Liu et al.2010; Comerford et al. 2011; Fu et al. 2012; Barrows et al.2012, 2013; McGurk et al. 2015; M¨uller-S´anchez et al.2015), single-peaked narrow AGN emission lines that ex-hibit a velocity offset relative to systemic can be usedto select offset AGN candidates (Comerford et al. 2009,2013). Follow-up observations of some of these galaxiesthat exhibit velocity-offset AGN emission lines in theirSloan Digital Sky Survey (SDSS) spectra have revealedthat the velocity-offset emission lines can be producedby offset AGNs or by shocked gas in outflows or inflows(Allen et al. 2015; Barrows et al. 2016; M¨uller-S´anchezet al. 2016).To determine the nature of SDSS J1354+1327, wehave observed the galaxy with the
Chandra X-rayObservatory , Hubble Space Telescope ( HST ) imaging,Apache Point Observatory Dual Imaging Spectrograph(APO/DIS) optical longslit spectroscopy, and adaptiveoptics-assisted Keck OH-Suppressing Infra-Red ImagingSpectrograph (Keck/OSIRIS) integral-field spectroscopy(IFS). We conclude that SDSS J1354+1327 is the bestobservational example of an AGN turning off and thenon, with a relic outflow to the south and a new AGNoutburst north of the galaxy center (Figure 1, left). Wefind that the velocity-offset emission lines are producedby shocks emanating from the new AGN outburst.The paper is organized as follows: in Section 2we present our observations, including space-based andground-based observations of SDSS J1354+1327 andanalyses of these data. In Section 3 we combine the ob-servations to interpret the nature of the emission sources a r X i v : . [ a s t r o - ph . GA ] O c t Comerford et al.in SDSS J1354+1327 and the nature of its companiongalaxy. In Section 4 we determine the nature of SDSSJ1354+1327 itself. In Section 5 we present our conclu-sions. We assume a Hubble constant H = 70 km s − Mpc − , Ω m = 0 .
3, and Ω Λ = 0 . OBSERVATIONS AND ANALYSIS2.1.
Optical SDSS Observations
The spectrum of SDSS J1354+1327 is classified as aType 2 AGN and has no evidence of broad emission lines(Oh et al. 2015). We obtain the fluxes of the emissionlines in the SDSS spectrum of the galaxy using the OSSYcatalog (Oh et al. 2011), which simultaneously fits an en-tire spectrum using stellar templates for the stellar kine-matics and Gaussian templates for the emission compo-nents. The host galaxy redshift, based on the templatefit to the stellar absorption features, is z = 0 . ± . III ] λ L [O III ] λ . × ergs − . The [O III ] λ . ± . − relative to the systemic velocity of thegalaxy’s stars (Comerford & Greene 2014). We find thatthe line flux ratios are log([O III ] λ / H β ) = 1 . ± .
05 and log([N II ] / H α ) = − . ± .
05, which placeSDSS J1354+1327 in the Seyfert region of the Baldwin-Phillips-Terlevich (BPT) diagram (Baldwin et al. 1981;Kewley et al. 2006).The companion galaxy SDSS J1354+1328 has a red-shift of z = 0 . ± . III ] λ L [O III ] λ . × erg s − . Theline flux ratios are log([O III ] λ / H β ) = 0 . ± . II ] / H α ) = − . ± .
08, which are in theSeyfert region of the BPT diagram.2.2.
Chandra/ACIS X-ray Observations
SDSS J1354+1327 was observed with
Chandra /ACISfor 9443 seconds on UT 2014 June 25, as part of theprogram GO4-15113X (PI: Comerford). The data weretaken with the telescope aimpoint on the ACIS S3 chip in“timed exposure” mode and telemetered to the groundin “faint” mode. We reduced the data with the latest
Chandra software (CIAO 4.6.1) and the most recent setof calibration files (CALDB 4.6.2).We used dmcopy to make sky images of the field inthe soft ( S , restframe 0 . − H , restframe2 −
10 keV) and total ( T , restframe 0 . −
10 keV) en-ergy ranges with pixels binned to 1/10th the native pixelsize. Then, we used
Sherpa to model the X-ray source asa two-dimensional Lorenztian function in beta2d . Ourmodel also included a background component of fixedcount rate, that we determined from an adjacent circu-lar region of 30 (cid:48)(cid:48) radius. We confirmed that the back-ground region does not contain any sources detected by wavdetect with a threshold of sigthresh = 10 − . Weset the initial position of the beta2d component to thelocation of the SDSS galaxy coordinates. After using psfSize to estimate that the radius of the point spreadfunction (PSF) is 1 . (cid:48)(cid:48)
07, we ran the fit in
Sherpa andallowed the model to fit a region around the galaxy co-ordinates of 3 times the PSF size at that location on thechip. We determined the best-fit model parameters by minimizing the Cash statistic using
Sherpa ’s implemen-tation of the ’Simplex’ minimization algorithm (Lagariaset al. 1998). To test for additional sources, we also at-tempted fitting a two-component beta2d model. How-ever, the amplitude of any second component is detectedat < σ significance, and therefore we adopt the single beta2d model.To determine the relative positions of the Chandra sources and the
HST sources, we registered the
Chan-dra and
HST /F160W images and estimated the relativeastrometric uncertainty between the two images. Due tothe small number of
Chandra sources and the relativelysmall
HST /F160W field of view, we could not directlyregister the two images. Instead, we registered bothimages with external images: SDSS ( u , g , r , i , and z )and the 2MASS point source catalog (Cutri et al. 2003).This provided us with six independent estimates of thetransformations and relative astrometric uncertainty be-tween the Chandra and
HST images. Finally, to mini-mize the relative astrometric uncertainty, we computedan error-weighted average of the six independent trans-formations. In the
Chandra physical ( X , Y ) coordinatesystem, the error-weighted averages of these six transfor-mations yield the final astrometric errors ∆ X = 0 . (cid:48)(cid:48) Y = 0 . (cid:48)(cid:48) HST coordinateframe, are (13:54:29.052, +13:27:56.70), (13:54:29.050,+13:27:56.81), and (13:54:29.047, +13:27:56.84), respec-tively. The errors on the positions are (0 . (cid:48)(cid:48)
19, 0 . (cid:48)(cid:48) . (cid:48)(cid:48)
07, 0 . (cid:48)(cid:48) . (cid:48)(cid:48)
06, 0 . (cid:48)(cid:48) BEHR ) code de-scribed in (Park et al. 2006).
BEHR takes as input theobserved soft and hard counts from both the source re-gion and a background region (which we measured us-ing calc data sum ). The code then uses a Bayesian ap-proach to estimate the expected values and uncertaintiesof the soft counts, hard counts, total counts, and hard-ness ratio. Using this approach, we found S = 13 . +3 . − . counts, H = 238 +13 . − . counts, and T = 252 +14 . − . counts.We then used Sherpa to model the unbinned energyspectrum of the extracted region over the observed en-ergy range of 2-8 keV. For our purposes, we are inter-ested in the observed and intrinsic fluxes integrated overrestframe soft, hard, and total energy ranges. There-fore, we fit the spectrum with a redshifted power law, F = E − Γ , (intended to represent the intrinsic AGN X-ray emission at the SDSS spectroscopic redshift, z SDSS ).We also included two multiplicative model componentsof absorbing neutral Hydrogen column densities. Thefirst absorbing component is fixed to the Galactic value, n H , gal , which we determined using an all-sky interpola-tion of the neutral Hydrogen in the Galaxy (Dickey &Lockman 1990). The second absorbing component is al-lowed to vary and assumed to be intrinsic to the source, n H , exgal , at the redshift z SDSS . All fluxes are k -corrected,with the observed values calculated from the model sum(including the absorbing components), while the intrin-sic values are calculated from the unabsorbed power lawn Active Galactic Nucleus Caught in the Act of Turning Off and On 3 Fig. 1.— (cid:48)(cid:48) × (cid:48)(cid:48) images of SDSS J1354+1327 and its companion galaxy SDSS J1354+1328, which is located 12.5 kpc to the northeast.Left: a four-color image, where the observations shown are HST
F160W (red), F606W (green), F438W (blue), and
Chandra restframe0 . −
10 keV (purple, one-twelfth size pixels smoothed with a 16 pixel radius Gaussian kernel). The inset shows a 1 . (cid:48)(cid:48) × . (cid:48)(cid:48) HST data shown to more clearly illustrate the northern bubble of ionized gas. Right: a V − H color map to illustrate the distribution of dust in the system. The color map is plotted on a logarithmic gray scale, where darkcorresponds to redder colors and light corresponds to bluer colors. In all images, north is up and east is to the left. component. Since we have used Cash statistics, we didnot subtract the background before modeling, but wehave confirmed that the counts over the 2-8 keV rangeare dominated by the source region, with a negligiblecontribution from the background region.For our first fit to the spectrum, we allowed Γ and n H , exgal to vary freely. We found a best-fit value ofΓ = − .
81 (and a corresponding upper limit on the col-umn density of 3 . × cm − ), which is not within thetypical range of observed power-law indices for AGNs,i.e. 1 ≤ Γ ≤ Sherpa ’s implementation of theLevenberg-Marquardt optimization method (Bevington1969). The best-fit column density is 1 . +0 . − . × cm − , where the errors do not account for the uncer-tainty in Γ, and the reduced Cash statistic is 0.94. Fig-ure 3 shows the best fit to the spectrum.The restframe hard absorbed (unabsorbed) X-rayfluxes are 1 . +0 . − . × − erg s − cm − (2 . +0 . − . × − erg s − cm − ) for 2-10 keV. We then used the redshift z SDSS to convert the X-ray fluxes to X-ray luminosities.The restframe hard absorbed (unabsorbed) luminositiesare L X, − = 1 . +0 . − . × erg s − (2 . +0 . − . × erg s − ).Finally, we searched for an X-ray source at the centerof the companion galaxy SDSS J1354+1328, but foundnone (detection of 0 . +0 . − . counts in the restframe0 . −
10 keV energy range). 2.3.
HST/WFC3 F160W, F606W, and F438WObservations
SDSS J1354+1327 was also observed with
HST /WFC3on UT 2014 April 28 (GO 13513, PI:Comerford), andthe observations covered three bands: UVIS/F438W ( B band, 954 seconds), UVIS/F606W ( V band, 900 sec-onds), and IR/F160W ( H band, 147 seconds). The HST field of view included SDSS J1354+1328, which is thecompanion galaxy to the northeast of SDSS J1354+1327.We used GALFIT V3.0 (Peng et al. 2010) to model thelight profiles of these two galaxies. GALFIT is capableof decomposing images of galaxies into multiple compo-nents, including a S´ersic profile (e.g., a galactic stellarbulge), exponential disk (e.g., a galactic disk), and animage PSF for bright unresolved sources (e.g., AGNs).We fit a S´ersic profile (plus a fixed, uniform sky compo-nent) to locate the position of each central stellar bulge,since S´ersic profiles have been empirically shown to begood approximations of the light profiles of stellar bulges(Graham & Driver 2005). To avoid modeling complexand irregular gas kinematics, we only ran GALFIT mod-els on the F160W image because it does not contain sig-nificant line emission from ionized gas and it is sensitiveto the central stellar bulge, which is the primary compo-nent of interest. As shown in Comerford et al. (2015),an additional unresolved point source is not necessaryin this model since the AGN contributes negligibly tothe F160W flux. The fit was run on a square region ofprojected physical size 40 kpc on each side, in order toexplore within 20 kpc of the AGN position.We find that the position of SDSS J1354+1327’s cen-tral stellar bulge is (13:54:29.045, +13:27:56.88), witherrors of (0 . (cid:48)(cid:48)
06, 0 . (cid:48)(cid:48) . . (cid:48)(cid:48) HST/F438W10 kpc8.24" HST/F438W1 kpc0.82" Chandra (2-10keV) HST +ChandraHST/F606W10 kpc8.24" HST/F606W1 kpc0.82" Chandra (2-10keV) HST +ChandraHST/F160W10 kpc8.24" HST/F160W1 kpc0.82" Chandra (2-10keV) HST +Chandra
Fig. 2.—
Top to bottom: F438W, F606W, and F160W observations of SDSS J1354+1327.
Left : HST image, with a box illustrating theregion of SDSS J1354+1327 shown in the figures to the right.
Middle left : 2 (cid:48)(cid:48) × (cid:48)(cid:48) HST image, with cyan crosses marking the centralemission source and the NE emission source (or in the case of F160W, marking only the stellar centroid).
Middle right : 2 (cid:48)(cid:48) × (cid:48)(cid:48) image ofrestframe 2 −
10 keV
Chandra observations, where the blue cross marks the position of the X-ray source and the size of the cross illustratesthe 3 σ errors on the X-ray source position, given the relative astrometric uncertainties between the Chandra image and the
HST bandshown.
Right : 2 (cid:48)(cid:48) × (cid:48)(cid:48) image of restframe 2 −
10 keV
Chandra observations, with
HST contours overlaid. The cyan and blue crosses areas shown in the middle left and middle right figures, respectively. In all panels, north is up and east is to the left.
Fig. 3.—
X-ray spectral fit to SDSS J1354+1327, where the toppanel shows the fit in red and the bottom panel shows the residuals. to each galaxy, we also find the H -band luminosity forSDSS J1354+1327 ( L H = 4 . × L (cid:12) ) and for thecompanion galaxy ( L H = 4 . × L (cid:12) ).To explore the distribution of dust in the galaxy, we made a V − H color map (Martini et al. 2003). Weconvolved the V and H images with a Gaussian kernelto put them at the same angular resolution, and thencomputed the magnitude difference V − H . The resultantcolor map is shown in Figure 1 (right). We find that thecenter of SDSS J1354+1327 is the most obscured part ofthe galaxy ( V − H = 2).Next, we turn our attention to the F438W andF606W observations, which cover a wavelength rangethat includes Hδ , Hγ , and [O III ] λ β , [O III ] λλ I ] λλ II ] λλ α (F606W). The F438Wand F606W observations are dominated by ionized gasemission and not stellar emission, since the opticallongslit observations show this to be the case for thesewavelength ranges and since the morphologies of theF438W and F606W emission are different from themorphology of the stellar continuum (traced in theF160W observations). We identify two main structuresin the F438W and F606W data: Spatially extended ionized gas to the south.
South ofthe galaxy center, there is a cone of clumpy, ionized gasn Active Galactic Nucleus Caught in the Act of Turning Off and On 5
Fig. 4.—
Segments of the two-dimensional APO/DIS longslit spectra taken at three different position angles: 0 (top; the companiongalaxy is visible above the target galaxy), 45 (middle), and 153 (bottom) degrees East of North. The blue channel spectra are shown on theleft, and the red channel spectra are shown on the right. The wavelengths shown are restframe wavelengths based on the galaxy’s stellarabsorption features. The longslit spectra are centered spatially on the galaxy continuum, with positive spatial positions to the north andnegative spatial positions to the south. (Figure 1, left). We map the gas detected in F606Wat > σ above the background, and we find that it isdistributed in a cone-like structure with an opening angleof 67 ◦ . The gas detected at > σ then extends ∼
10 kpcradially from the galaxy center. The range of colors inthe cone is 0 . < V − H < .
2, and the patchiness of thecolor indicates that in some regions the far side of thecone may be obscured. We explore this in more detail inSection 3.2.
Northern bubble.
North of the galaxy center, there isa loop of emission that extends to a maximum distanceof 0 . (cid:48)(cid:48) V − H = 0 .
7. By fitting a S´ersic component, wefind that the brightest component of this bubble is lo-cated northeast of the galaxy center. We call this featurethe NE source, and in F606W its apparent magnitude is21.4 (compared to an apparent magnitude of 18.7 forthe galaxy). The positions of the NE source in F438W(13:54:29.064, +13:27:57.10) and F606W (13:54:29.062,+13:27:57.10) agree to within 0 . (cid:48)(cid:48)
04, so we take the NEsource position to be the average of the positions mea-sured in F438W and F606W. The NE source is located0 . (cid:48)(cid:48)
33 (0.40 kpc) from the galaxy center. 2.4.
APO/DIS Optical Longslit Observations
We observed SDSS J1354+1327 with APO/DIS on UT2015 December 18 and UT 2016 June 30. We used a1200 lines mm − grating, centered so that the wave-length range covered H β and [O III ] on the blue channeland [N II ] and H α on the red channel. We used a 1 . (cid:48)(cid:48) . (cid:48)(cid:48)
42 on the blue chan-nel and 0 . (cid:48)(cid:48)
40 on the red channel. Following the approachused to study active galaxies with double-peaked narrowemission lines (Comerford et al. 2012; Nevin et al. 2016),we observed SDSS J1354+1327 at multiple different slitposition angles to understand the kinematics of the fea-tures of the galaxy that are seen in the
HST images. Weobserved the galaxy at three different slit position angles(given in degrees East of North): 0 ◦ to cover the emis-sion seen in the north-south direction and the companiongalaxy to the north, 153 ◦ to cover the bright clumps ofemission seen to the south of the galaxy center, and 45 ◦ to cover the NE emission source. We observed for 1200sat each position angle, and the data were reduced usingstandard IRAF procedures.The longslit spectra are shown in Figure 4. We fitGaussians to the [O III ] λ α flux at each po-sition along the slit where the emission line is detected Comerford et al. Fig. 5.—
Line-of-sight velocity offsets of emission lines along thethree different position angles used for the APO/DIS longslit obser-vations. H α is illustrated with filled red circles, while [O III ] λ with a signal-to-noise ratio >
10, and we extracted theline-of-sight velocities. These are shown in Figure 5.Next we identify where the peak of [O
III ] λ III ] λ . × − erg s − cm − , which is twice the integrated fluxat position angle 0 and 2.4 times the integrated flux atposition angle 153), and we fit a Gaussian to the spatialdistribution of [O III ] λ III ] λ . (cid:48)(cid:48) ± . (cid:48)(cid:48)
08 to the north of the galaxy along position angle45, which is spatially coincident with the NE source seenin the
HST images. The velocity of the [O
III ] λ . ± . − .This explains the velocity offset of 69 . ± . − seen in the SDSS spectrum of SDSS J1354+1327.We then use emission line flux ratios in the BPT dia-gram (Baldwin et al. 1981; Veilleux & Osterbrock 1987;Kewley et al. 2006) to identify the sources of the ion-ized emission in SDSS J1354+1327. We measure theseline flux ratios at each position along the slit where thesignal-to-noise ratio is > I ] λ α and He II λ β at each positionalong the slit where the signal-to-noise ratio is > CLOUDY (Ferland 1996) and the shock models are computed with
MAPPINGSIII (Dopita & Sutherland 1996). In order to explore the densities of the materialwithin the galaxy, we also measure the line flux ratio[S II ] λλ > ∼ < n e (cm − ) ∼ < ; Osterbrock & Ferland 2006), and that theelectron density is higher south of the galaxy center andlower north of the galaxy center.From the longslit spectra, we identify three main kine-matic components in the galaxy: Rotating disk . The rotating disk is most apparent atposition angle 45 (Figure 5, middle), where the velocitiesprogress from redshifts to the south to blueshifts to thenorth. We find that the H α emission is best fit by arotating disk that has a position angle on the sky of 53 ◦ East of North, is inclined 50 ◦ into the plane of the sky,has an intrinsic velocity of 182 km s − , has a radius of2.5 kpc, and whose kinematic center is located at thegalaxy’s stellar centroid. This rotating disk is also seenin Pa α , with the same disk parameters. Spatially extended ionized gas to the south . South ofthe galaxy center, we find ionized gas that extends todistances (cid:38)
III ] λ α is contaminated by thedisk rotation. We find that the [O III ] λ − red-shifted relative to systemic (along position angle 153) to20 km s − blueshifted relative to systemic (along posi-tion angle 153; Figure 5). The line flux ratios indicatea photoionized origin for emission south of the galaxycenter (Figure 7). Northern bubble.
North of the galaxy center there isionized gas that extends ∼ III ] λ − redshifted relative to systemic(along position angle 45) to 20 km s − blueshifted rela-tive to systemic (along position angle 0; Figure 5). The[O III ] λ HST ob-servations. The line flux ratios at the position of the NEsource are inconsistent with pure photoionization mod-els, and the high He II λ β ratio can instead beexplained by a shock with a precursor H II region (e.g.,Sutherland et al. 1993); Moy & Rocca-Volmerange (2002)find regions of AGN emission with similar line flux ratiosand interpret them also as produced by a shock with aprecursor.2.5. Keck/OSIRIS Near-infrared IFS Observations
We observed SDSS J1354+1327 on UT 2013 April 3using Laser Guide Star Adaptive Optics (LGS-AO) withKeck/OSIRIS integral-field spectroscopy. We used theKn1 filter with the 0 . (cid:48)(cid:48) . (cid:48)(cid:48) × . (cid:48)(cid:48) (cid:48)(cid:48) diameter SDSS fiber.The observations covered an observed wavelength rangeof 1.955 to 2.055 µ m and detected the Pa α µ m emis-sion line, which traces the spatial distribution and kine-matics of the ionized gas in the galaxy. We used then Active Galactic Nucleus Caught in the Act of Turning Off and On 7 log [NII] λ α l o g [ O III ] λ / H β PA = 0 D i s t a n c e [ ] ( N e g a t i v e i s S ) log [NII] λ α l o g [ O III ] λ / H β PA = 45 D i s t a n c e [ ] ( N e g a t i v e i s S ) log [NII] λ α l o g [ O III ] λ / H β PA = 153 D i s t a n c e [ ] ( N e g a t i v e i s S ) Fig. 6.—
BPT diagrams for the three optical longslit position angles: 0 (left), 45 (middle), and 153 (right) degrees East of North.The solid line shows the theoretical maximum for starbursts (Kewley et al. 2001), the dashed line shows the division between purelystar-forming and composite (contribution from AGN and star formation) galaxies (Kauffmann et al. 2003), and the dotted line shows thedivision between Seyferts and LINERs (Kewley et al. 2006). Colored data points and 1 σ error bars are plotted for each position along theslit, where zero is defined as the position of the host galaxy continuum on the slit. Fig. 7.— [O I ] λ α vs. He II λ β . The orangepoint illustrates the line flux ratios of the emission located 3 kpcsouth of the galaxy center along position angle 153 ◦ , which is partof the spatially extended ionized gas to the south. The green pointillustrates the line flux ratios of the emission located 0.5 kpc northof the galaxy center along position angle 45 ◦ , at the location of theNE source. The solid line represents pure photoionization modelswith ionization parameter U varying between log U = − U = − α = −
1. The dashed line represents a model of shocks anda matter bound precursor, where the shock has v = 1000 km s − (models adapted from Moy & Rocca-Volmerange 2002). galaxy nucleus as the tip-tilt star, observed at a positionangle of 0 degrees East of North, and integrated for 30minutes. The data were reduced with the OSIRIS datareduction pipeline, and we used the IDL code LINEFIT(Davies et al. 2007) to create 2D kinematic and flux dis-tribution maps (Figure 9).From the OSIRIS observations, we identify three mainkinematic components in the galaxy: Rotating disk . To model the rotating disk, we firstmask the pixels in the Pa α velocity map that have a fluxdensity lower than 10% of the peak of the continuumemission. We then use the KINEMETRY code (Krajnovi´cet al. 2006) to model the rotating disk structure seenextending from the northeast (blueshifted) to the south-west (redshifted) in Pa α . The parameters of the best-fitmodel are a position angle of 55 degrees East of North,an inclination of 53 degrees, and a kinematic center that Fig. 8.—
Electron density measured along the APO/DIS longslitspectrum taken at position angle 45 ◦ east of north. Zero is definedas the position of the host galaxy continuum on the slit, positivespatial positions are to the north, and negative spatial positions areto the south. Along this position angle, the NE source is located0 . (cid:48)(cid:48)
41 to the north (this position corresponds with the first datapoint north of the continuum), and the electron density is an orderof magnitude lower at the NE source than at the southern positionson the slit. is spatially offset by 0 . (cid:48)(cid:48) α are consistent with those of the rotat-ing disk observed in the optical longslit spectra. Afterthe disk model is subtracted from the data, we exam-ine the velocity residuals, which have ±
20 km s − errors(Figure 9, bottom right). The velocity residuals are notsignificant enough to fit a secondary component. We findthat the kinematic center of the rotating disk is offset by0 . (cid:48)(cid:48) Spatially extended ionized gas to the south . While theOSIRIS field of view is dominated by the rotating disk,there is also extraplanar Pa α emission visible to the southof the galaxy center, and this emission is detected at 5 σ significance out to the edge of the field of view (Figure 9,top middle). The line-of-sight velocities of the gas are ±
30 km s − . The Pa α emission also follows the conicalstructure seen in the HST observations. Comerford et al.
Fig. 9.—
Two-dimensional maps of the K -band continuum (top left), Pa α flux distribution (top middle), and Pa α velocity (top right)for SDSS J1354+1327, where the data have been smoothed with a median filter of 3 pixels. The contours delineate the K -band continuumemission in the top left and middle panels, and the ionized gas morphology in the top right panel. The center of the galaxy (the peak ofcontinuum emission at 1.9 µ m) is located at position (0,0). Regions in white in the top right panel correspond to pixels where the lineproperties are uncertain and thus were masked out. These rejected pixels are those with a flux density lower than 5% of the peak of Pa α emission. Bottom row, from left to right: Pa α velocity map (where pixels that have a flux density lower than 10% of the peak of thecontinuum emission have been masked), velocity map of the rotating disk that is the best fit to the data, and the velocity residuals (notethe different scale for the velocity color bar). Each map is centered on the kinematic center of the galaxy. In all maps north is up and eastis to the left. Northern bubble . There is spatially extended Pa α emis-sion associated with the loop seen in the HST data. RESULTS3.1.
Nature of the X-ray Source: A Central AGN
The X-ray source in SDSS J1354+1327 has an observedluminosity of L X, − = 1 . × erg s − , which isbrighter than the typical AGN luminosity threshold of L X, − > erg s − and shows that the X-raysource is associated with an AGN. Mid-infrared colorsfrom the Wide-field Infrared Survey Explorer ( WISE )also show that an AGN is present (SDSS J1354+1327has W − W . ± .
03, where W − W ≥ . n H,exgal = 2 × cm − , which indicatesthat the source is an X-ray absorbed AGN. This is con-sistent with the V − H dust map, which shows that thevery center of the galaxy is the most obscured (Figure 1,right).Next, we consider the position of the X-ray sourcewithin the host galaxy. We have identified the positionof the X-ray source, the positions of the central stellarnucleus of the galaxy and the NE emission source vis-ible in the HST
F438W and F606W observations, andperformed astrometric corrections. We compare the po-sitions to determine if the X-ray AGN position alignswith any of these galaxy features (Figure 2, right). Ifthe X-ray AGN is associated with the NE source, thenthe AGN is located 0.40 kpc from the center of SDSSJ1354+1327 and this spatial offset could be explained asan offset AGN, a gravitational recoil AGN, or a gravita- tional slingshot AGN. We find that the position of theX-ray AGN is more consistent with the position of thestellar centroid of the galaxy (consistent to within (cid:46) σ )than with the position of the NE source (consistent towithin (cid:46) . σ ). Consequently, we conclude that the X-ray AGN is most likely located at the galaxy center.3.2. Nature of the Spatially Extended Ionized Gas tothe South: A Photoionized AGN Outflow
There is extraplanar ionized gas extending south of thegalaxy center, as seen in the
HST
F606W and F438W ob-servations, the APO/DIS optical longslit observations,and the Keck/OSIRIS near-infrared IFS observations.The gas is clumpy, as seen morphologically in the
HST data (Figure 1, left) and in the mixture of redshifted andblueshifted line-of-sight velocities of the emission linesobserved with APO/DIS (Figure 5).Here, we aim to determine whether the gas is inflow-ing, outflowing, or passively photoionized by the centralAGN. If the AGN is passively photoionizing the gas inthe galaxy, then there should be symmetric photoion-ized cones on either side of the AGN due to the colli-mating torus (e.g., Schmitt et al. 2003), but this is notthe case for SDSS J1354+1327. In the passive photoion-ization scenario, the gas also should be rotating (withpossible deviations due to the interaction with the com-panion galaxy) since it is following the galaxy potential.We fit a rotating disk to the optical and near-infraredspectra, and we find that the rotating disk is orientedwith a position angle 54 degrees East of North, but norotation is seen in the gas extending to the south of thatdisk. Instead the southern gas has blueshifts and red-shifts observed along directions not coincident with theaxis of rotation, as can be seen in the velocities in then Active Galactic Nucleus Caught in the Act of Turning Off and On 9longslit data. This is evidence of an inflow or outflow,rather than passively photoionized gas.The morphology of the southern gas distinguishes be-tween the inflow and outflow scenarios. As shown in the
HST images (Figure 1, left), the gas has a conical mor-phology with the cone beginning at the galaxy center andextending to the south, outside the plane of rotation ofthe galaxy. This conical morphology is typical of AGNoutflows (e.g., Mulchaey et al. 1996; Schmitt et al. 2003),where the torus provides the collimation (e.g., Antonucci& Miller 1985; Malkan et al. 1998). In contrast, coni-cal morphologies are not expected for inflows, which aretypically radial streamers of gas (e.g., Iono et al. 2004;M¨uller S´anchez et al. 2009).We conclude that the spatially extended ionized gas tothe south is an outflow. We apply an analytic MarkovChain Monte Carlo model to the longslit observationsto model the data as a biconical outflow, as done forsimilar longslit observations of galaxies in Nevin et al.(2017). We find that the data are best fit (with a reduced χ = 1 .
5) by a bicone extending south of the galaxycenter at an inclination of 27 +14 − degrees out of the planeof the sky, a position angle of 190 +26 − degrees east of northon the plane of the sky, a half opening angle of 37 +10 − degrees, a turnover radius of 1 . +2 . − . kpc, a maximumvelocity at the turnover radius V max = 40 +40 − km s − ,and a lateral surface area A = 4 +14 − kpc . The observedblueshifted and redshifted velocities of the gas imply thatwe are seeing parts of the front and rear facing walls ofthe outflow (e.g., Bae & Woo 2016), which fits with theclumpiness of the gas and dust partially obscuring theview (Figure 1, right). When we project the cone ontothe plane of the sky, we find an observed opening angleof 66 +18 − degrees, which is consistent with the openingangle of 67 degrees measured from the HST data.Then, we use the best-fit parameters of the outflowmodel, our measurement of the electron density n e , andassume a filling factor f = 0 .
01 to be conservative inour energy calculation to estimate the mass outflow rate˙ M = m p n e V max f A , where m p is the proton mass. Wefind ˙ M = 6 . +52 . − . M (cid:12) yr − .This outflow could be driven by either star formationor an AGN. First we consider the scenario of a star forma-tion driven outflow. To set an upper limit on the star for-mation rate (SFR) in SDSS J1354+1327, we assume thatall of the H α emission is associated with star formation.Using the luminosity measured from the SDSS spectrum, L H α = 1 . × erg s − , the SFR from Kennicutt& Evans (2012) is then SF R = 5 . × − L H α =0 . M (cid:12) yr − . When we convert this SFR to a mass out-flow rate we find an upper limit of ˙ M SF = 0 . M (cid:12) yr − (Veilleux et al. 2005), which is too weak to drive the ob-served outflow. We also note that a star formation drivenoutflow that is powered by supernovae in the disk of thegalaxy typically has a wide, chimney-shaped morphologyof the gas being driven from the disk into the halo (e.g.,Norman & Ikeuchi 1989), which is inconsistent with theconical morphology of the ionized gas observed in SDSSJ1354+1327.Consequently, we conclude that the outflow is AGNdriven. This is consistent with the [O III ] λ β and [N II ]/H α line flux ratios at the location of the outflow,which show that the emission is Seyfert driven. Further,the [O I ] λ α and He II λ β line flux ratiosand the conical morphology each indicate that the emis-sion is photoionized (e.g., Wilson & Tsvetanov 1994).The AGN luminosity is also sufficient for photoionizingthe cone of gas, based on photoionization models com-puted using the spectral synthesis code CLOUDY (Ferlandet al. 2013; Richardson et al. 2014). We conclude thatthe spatially extended ionized gas to the south is a pho-toionized AGN outflow.Although we do not know the velocity of the outflowat the time it was launched, we can use the current ob-served velocity and spatial extent of the outflow to findan upper limit of 10 yrs on the timescale for the outflowto produce the cone of gas. The light-travel timescaleplus the recombination timescale ( t rec ≈ ( αn e ) − , wherethe recombination coefficient α = 1 . × − cm s − for [O III ] and n e is from Section 2.4; Osterbrock & Fer-land 2006) to illuminate the gas is then ∼
50 yrs, whichis negligible. Given that the emission can be observed forup to 10 yrs (e.g., Schawinski et al. 2015), we concludethat the outflow was likely launched ∼ < yrs ago.3.3. Nature of the Northern Bubble: A Shocked AGNOutflow
The
HST
F438W and F606W observations reveal aloop of gas extending 0.7 kpc north of the galaxy center,as well as a NE source that is located 0.40 kpc northeastof the galaxy center and is embedded in the loop of emis-sion (Figure 1, left). The peak of [O
III ] λ III ] λ ∼ < yrs ago. Taking intoaccount the length of time that the emission signaturescan linger, the outflow must have been launched ∼ < Fig. 10.—
Left: 3D drawing of the main components of SDSS J1354+1327: a southern conical outflow (with clumpiness shown), anorthern bubble, and a rotating disk. Arrows show the directions of the gas motion in each component, and an eye shows the orientationof the observer. Right: face on view of the galaxy’s main components, as shown from the perspective of the observer. yrs ago.3.4.
Companion Galaxy to SDSS J1354+1327
SDSS J1354+1327 has a companion galaxy, SDSSJ135429.17+132807.3, that is located 12.5 kpc to thenortheast and is redshifted by 76 km s − relative to SDSSJ1354+1327. Emission line diagnostics of the compan-ion’s SDSS spectrum suggest that it hosts an AGN, andLiu et al. (2011) also noted that SDSS J1354+1327 andSDSS J1354+1328 are a pair of active galaxies. However,there is no WISE source detected at the location of thecompanion galaxy. Further, we do not detect an AGN inour
Chandra observations of the companion galaxy.The lack of an X-ray detection suggests that the AGNis obscured, the AGN’s X-ray luminosity is low on thepredicted [O
III ] λ III ] λ III ] λ III ] λ β = 0 .
44) could be produced bya starburst instead of an AGN, given that models predictthat starbursts can produce line ratios that high. Giventhat the dust map indicates obscuration at the galaxycenter ( V − H = 2 . HST observations show tidal tails of stars con-necting the primary and companion galaxies, indicatingthat they are already interacting (Figure 2, bottom left).Taking the mass ratio to be the luminosity ratio of thetwo stellar bulges, the merger’s mass ratio is 0 .
98 (SDSSJ1354+1327 is slightly less massive). NATURE OF SDSS J1354+1327: EPISODIC AGNOUTBURSTSWe have shown that SDSS J1354+1327 hosts asym-metric AGN-driven outflows. South of the galaxy centerthere is a large, ∼
10 kpc extended outflow of ionized gas,while north of the galaxy center there is a small scale( ∼ n e = 45 cm − ) is an order ofmagnitude lower than the densities associated with thesouthern outburst ( n e ≥
420 cm − ), which does not sup-port the concurrent AGN outburst scenario.Another possible explanation for the asymmetric out-flows in SDSS J1354+1327 is discontinuous AGN out-burst activity, which has been seen in simulations andcan create outbursts on opposite sides of the galaxy suchas those seen in SDSS J1354+1327 (Hopkins & Quataert2010; Gabor & Bournaud 2014). Clumpy structure inthe ISM can lead to highly variable accretion rates ontothe supermassive black hole (e.g., Hickox et al. 2014),where each high-accretion event produces an AGN out-burst. This outburst injects thermal energy into the am-bient gas, launching an outflow. Initially, the heated gasencounters and shocks cold gas, making shocks an indica-tor of a recent AGN outburst (Gabor & Bournaud 2014).n Active Galactic Nucleus Caught in the Act of Turning Off and On 11After this initial period of fast expansion and shocks, theoutflow slows and photoionization of the surrounding gasbecomes the dominant emission source. Consequently,two sequential AGN outbursts can produce asymmetricoutflows; the first AGN outburst leaves behind a largephotoionized cone of emission, while the later AGN out-burst produces small-scale shock emission.The observations support this picture of two sequen-tial AGN outbursts. We interpret the southern outflowas the result of an older AGN outburst, based on thephotoionized nature of the gas (indicated by the opticalline flux ratios and the gas’s conical morphology) andthe large spatial extent ( ∼
10 kpc) of the gas. We inter-pret the northern outflow as a new outflow that is beingdriven by a recent AGN outburst, based on the shockednature of the gas (indicated by the optical line flux ra-tios, the off-nuclear peak in [O
III ] λ ∼ z = 0 .
12 galaxy SDSS J1356+1026.SDSS J1356+1026 also has velocity-offset emission linesin its SDSS spectrum (although they are double peaked,instead of single peaked as for SDSS J1354+1327), large-scale AGN-driven outflows, and a companion galaxy lo-cated 3 kpc away (Liu et al. 2010; Fu et al. 2012; Greeneet al. 2012, 2014; Comerford et al. 2015). The majordifference is that SDSS J1356+1026 is much more lu-minous (bolometric luminosity 1 . × erg s − ) andpowerful: SDSS J1356+1026 is an ultra-luminous in-frared galaxy hosting a quasar that is powering large-scale ( ∼
10 kpc) symmetric ∼ − outflows ofionized gas, and it also hosts a compact ( ∼ . (cid:46) yrs. CONCLUSIONSWe have used
Chandra , HST , APO/DIS opticallongslit spectroscopy, and Keck/OSIRIS near-infraredAO-assisted IFS to determine the nature of the z = 0 . − . We determinedthat the velocity offset arises from the peak in emissionline gas, which is located 0.40 kpc northeast of the galaxycenter.We conclude that SDSS J1354+1327 is a galaxy whoseAGN has recently turned on and then on again, throughtwo accretion events that led to two sequential AGN out-bursts. The two accretion events arose from clumpy ISMbeing driven to the center of SDSS J1354+1327 as a re-sult of a tidal interaction with the companion galaxy lo-cated 12.5 kpc away. The outflow visible to the south ofthe galaxy center is remnant emission lingering from apast AGN outburst, because it is photoionized gas witha large spatial extent of the gas ( ∼
10 kpc). The emis-sion north of the galaxy center is a new AGN outburstthat launched a new outflow, based on the shocked na-ture of the gas and the smaller spatial extent of the gas( ∼ (cid:46) yrs.SDSS J1354+1327 is the strongest observational ex-ample of a galaxy with an AGN that has turned offand then on again. As such, SDSS J1354+1327 fitsinto the broader picture of episodic AGN activity thatalso includes many observations of light echoes of pastAGN outbursts. Our attention was initially drawn toSDSS J1354+1327 due to its narrow emission lines thatare offset in line-of-sight velocity from systemic. Thesevelocity-offset emission lines are commonly indicators ofshocks, and one mechanism that produces shocks is a re-cent AGN outburst being driven into the ISM. Therefore,future analyses of galaxies with velocity-offset AGN emis-sion lines will certainly uncover young AGN outburstsand some fraction of them may have evidence of pastAGN activity as well.J.M.C. thanks Nadia Zakamska for useful discussions.We also thank the anonymous referee for comments thathave improved the clarity of this paper. Support for thiswork was provided by NASA through Chandra AwardNumber GO4-15113X issued by the Chandra X-ray Ob-servatory Center, which is operated by the Smithsonian2 Comerford et al.Astrophysical Observatory for and on behalf of NASAunder contract NAS8-03060. Support for HST programnumber GO-13513 was provided by NASA through agrant from the Space Telescope Science Institute, whichis operated by the Association of Universities for Re-search in Astronomy, Inc., under NASA contract NAS5-26555.The scientific results reported in this article are basedin part on observations made by the Chandra X-ray Ob-servatory, and this research has made use of softwareprovided by the Chandra X-ray Center in the applica-tion packages CIAO, ChIPS, and Sherpa. The resultsreported here are also based on observations made withthe NASA/ESA Hubble Space Telescope, obtained at theSpace Telescope Science Institute, which is operated bythe Association of Universities for Research in Astron-omy, Inc., under NASA contract NAS 5-26555. Theseobservations are associated with program number GO-13513. Some of the observations reported here were obtainedat the Apache Point Observatory 3.5m telescope, whichis owned and operated by the Astrophysical ResearchConsortium.Some of the data presented herein were obtained at theW.M. Keck Observatory, which is operated as a scientificpartnership among the California Institute of Technol-ogy, the University of California and the National Aero-nautics and Space Administration. The Observatory wasmade possible by the generous financial support of theW.M. Keck Foundation. The authors wish to recognizeand acknowledge the very significant cultural role andreverence that the summit of Mauna Kea has alwayshad within the indigenous Hawaiian community. We aremost fortunate to have the opportunity to conduct ob-servations from this mountain. Facilities: CXO (ACIS),
HST (WFC3), APO (DualImaging spectrograph), Keck:I (OSIRIS)
REFERENCESAllen, J. T., Schaefer, A. L., Scott, N., Fogarty, L. M. R., Ho, I.-T.,Medling, A. M., Leslie, S. K., Bland-Hawthorn, J., Bryant, J. J.,Croom, S. M., Goodwin, M., Green, A. W., Konstantopoulos,I. S., Lawrence, J. S., Owers, M. S., Richards, S. N., & Sharp,R. 2015, MNRAS, 451, 2780Antonucci, R. R. J., & Miller, J. S. 1985, ApJ, 297, 621Bae, H.-J., & Woo, J.-H. 2016, ApJ, 828, 97Baldwin, J. A., Phillips, M. M., & Terlevich, R. 1981, PASP, 93, 5Barrows, R. S., Comerford, J. M., Greene, J. E., & Pooley, D. 2016,ApJ, 829, 37Barrows, R. S., Sandberg Lacy, C. H., Kennefick, J., Comerford,J. M., Kennefick, D., & Berrier, J. C. 2013, ApJ, 769, 95Barrows, R. S., Stern, D., Madsen, K., Harrison, F., Assef, R. J.,Comerford, J. M., Cushing, M. C., Fassnacht, C. D., Gonzalez,A. H., Griffith, R., Hickox, R., Kirkpatrick, J. D., & Lagattuta,D. J. 2012, ApJ, 744, 7Bevington, P. R. 1969, Data reduction and error analysis for thephysical sciencesClark, N. E., Tadhunter, C. N., Morganti, R., Killeen, N. E. B.,Fosbury, R. A. E., Hook, R. N., Siebert, J., & Shaw, M. A. 1997,MNRAS, 286, 558Comerford, J. M., Gerke, B. F., Newman, J. A., Davis, M., Yan,R., Cooper, M. C., Faber, S. M., Koo, D. C., Coil, A. L., Rosario,D. J., & Dutton, A. A. 2009, ApJ, 698, 956Comerford, J. M., Gerke, B. F., Stern, D., Cooper, M. C., Weiner,B. J., Newman, J. A., Madsen, K., & Barrows, R. S. 2012, ApJ,753, 42Comerford, J. M., & Greene, J. E. 2014, ApJ, 789, 112Comerford, J. M., Pooley, D., Barrows, R. S., Greene, J. E.,Zakamska, N. L., Madejski, G. M., & Cooper, M. C. 2015, ApJ,806, 219Comerford, J. M., Pooley, D., Gerke, B. F., & Madejski, G. M.2011, ApJ, 737, L19+Comerford, J. M., Schluns, K., Greene, J. E., & Cool, R. J. 2013,ApJ, 777, 64Cutri, R. M., Skrutskie, M. F., van Dyk, S., Beichman, C. A.,Carpenter, J. M., Chester, T., Cambresy, L., Evans, T., Fowler,J., Gizis, J., Howard, E., Huchra, J., Jarrett, T., Kopan, E. L.,Kirkpatrick, J. D., Light, R. M., Marsh, K. A., McCallon, H.,Schneider, S., Stiening, R., Sykes, M., Weinberg, M., Wheaton,W. A., Wheelock, S., & Zacarias, N. 2003, VizieR Online DataCatalog, 2246Davies, R. I., Mueller S´anchez, F., Genzel, R., Tacconi, L. J., Hicks,E. K. S., Friedrich, S., & Sternberg, A. 2007, ApJ, 671, 1388DeGraf, C., Dekel, A., Gabor, J., & Bournaud, F. 2017, MNRAS,466, 1462Dickey, J. M., & Lockman, F. J. 1990, ARA&A, 28, 215Dopita, M. A., & Sutherland, R. S. 1996, ApJS, 102, 161Fabian, A. C. 2012, ARA&A, 50, 455Faucher-Gigu`ere, C.-A., & Quataert, E. 2012, MNRAS, 425, 605Ferland, G. J. 1996, Hazy, A Brief Introduction to Cloudy 90Ferland, G. J., Porter, R. L., van Hoof, P. A. M., Williams, R. J. R.,Abel, N. P., Lykins, M. L., Shaw, G., Henney, W. J., & Stancil,P. C. 2013, RMXAA, 49, 137Fischer, T. C., Crenshaw, D. M., Kraemer, S. B., Schmitt, H. R.,Mushotsky, R. F., & Dunn, J. P. 2011, ApJ, 727, 71 F¨orster Schreiber, N. M., Genzel, R., Newman, S. F., Kurk, J. D.,Lutz, D., Tacconi, L. J., Wuyts, S., Bandara, K., Burkert, A.,Buschkamp, P., Carollo, C. M., Cresci, G., Daddi, E., Davies, R.,Eisenhauer, F., Hicks, E. K. S., Lang, P., Lilly, S. J., Mainieri,V., Mancini, C., Naab, T., Peng, Y., Renzini, A., Rosario, D.,Shapiro Griffin, K., Shapley, A. E., Sternberg, A., Tacchella, S.,Vergani, D., Wisnioski, E., Wuyts, E., & Zamorani, G. 2014,ApJ, 787, 38Fu, H., Yan, L., Myers, A. D., Stockton, A., Djorgovski, S. G.,Aldering, G., & Rich, J. A. 2012, ApJ, 745, 67Gabor, J. M., & Bournaud, F. 2013, MNRAS, 434, 606—. 2014, MNRAS, 441, 1615Graham, A. W., & Driver, S. P. 2005, PASA, 22, 118Greene, J. E., Pooley, D., Zakamska, N. L., Comerford, J. M., &Sun, A.-L. 2014, ApJ, 788, 54Greene, J. E., Zakamska, N. L., & Smith, P. S. 2012, ApJ, 746, 86Hickox, R. C., Mullaney, J. R., Alexander, D. M., Chen, C.-T. J.,Civano, F. M., Goulding, A. D., & Hainline, K. N. 2014, ApJ,782, 9Hopkins, P. F., & Quataert, E. 2010, MNRAS, 407, 1529Iono, D., Yun, M. S., & Mihos, J. C. 2004, ApJ, 616, 199Ishibashi, W., & Courvoisier, T.-L. 2010, Astronomy &Astrophysics, 512, A58Kauffmann, G., Heckman, T. M., Tremonti, C., Brinchmann, J.,Charlot, S., White, S. D. M., Ridgway, S. E., Brinkmann,J., Fukugita, M., Hall, P. B., Ivezi´c, ˇZ., Richards, G. T., &Schneider, D. P. 2003, MNRAS, 346, 1055Keel, W. C., Maksym, W. P., Bennert, V. N., Lintott, C. J.,Chojnowski, S. D., Moiseev, A., Smirnova, A., Schawinski, K.,Urry, C. M., Evans, D. A., Pancoast, A., Scott, B., Showley, C.,& Flatland, K. 2015, AJ, 149, 155Kennicutt, R. C., & Evans, N. J. 2012, ARA&A, 50, 531Kewley, L. J., Dopita, M. A., Sutherland, R. S., Heisler, C. A., &Trevena, J. 2001, ApJ, 556, 121Kewley, L. J., Groves, B., Kauffmann, G., & Heckman, T. 2006,MNRAS, 372, 961Koyama, K., Maeda, Y., Sonobe, T., Takeshima, T., Tanaka, Y.,& Yamauchi, S. 1996, PASJ, 48, 249Krajnovi´c, D., Cappellari, M., de Zeeuw, P. T., & Copin, Y. 2006,MNRAS, 366, 787Lagarias, J. C., Reeds, J. A., Wright, M. H., & Wright, P. E. 1998,SIAM Journal on optimization, 9, 112LaMassa, S. M., Cales, S., Moran, E. C., Myers, A. D., Richards,G. T., Eracleous, M., Heckman, T. M., Gallo, L., & Urry, C. M.2015, ApJ, 800, 144Lintott, C. J., Schawinski, K., Keel, W., van Arkel, H., Bennert,N., Edmondson, E., Thomas, D., Smith, D. J. B., Herbert, P. D.,Jarvis, M. J., Virani, S., Andreescu, D., Bamford, S. P., Land,K., Murray, P., Nichol, R. C., Raddick, M. J., Slosar, A., Szalay,A., & Vandenberg, J. 2009, MNRAS, 399, 129Liu, X., Shen, Y., Strauss, M. A., & Greene, J. E. 2010, ApJ, 708,427Liu, X., Shen, Y., Strauss, M. A., & Hao, L. 2011, ApJ, 737, 101Malkan, M. A., Gorjian, V., & Tam, R. 1998, ApJS, 117, 25Martini, P., Regan, M. W., Mulchaey, J. S., & Pogge, R. W. 2003,ApJS, 146, 353McGurk, R. C., Max, C. E., Medling, A. M., Shields, G. A., &Comerford, J. M. 2015, ApJ, 811, 14Moy, E., & Rocca-Volmerange, B. 2002, A&A, 383, 46 n Active Galactic Nucleus Caught in the Act of Turning Off and On 13n Active Galactic Nucleus Caught in the Act of Turning Off and On 13