Understanding active galactic nuclei (AGN) feedback is essential for building a coherent picture of the evolution of the super massive black hole and its host galaxy. To that end we have analysed the inner kiloparsec of a sample of 30 local AGN with spatially resolved optical spectroscopy. In this talk I will review the analysis of the ionised gas for the galaxies in our sample, including kinematical maps, emission line ratios and fluxes. The W80 kinematical index is used to trace outflows, and also to provide an estimate for the outflowing velocity. Electron densities, derived from the [S II] ΛΛ6716, 6731Å lines, along with Hα luminosities and the sizes of the outflowing regions are employed in estimates of the outflowing gas mass. We find a median mass outflow rate of Ṁ = 0.3 M⊙ yr-1 and median outflow power of log [P/(erg s-1)] = 40.4.

We investigate the processes of active galactic nuclei (AGN) feeding and feedback in the narrow line regions (NLRs) and host galaxies of nearby AGN through spatially resolved spectroscopy with the Gemini Near-Infrared Integral Field Spectrograph (NIFS) and the Hubble Space Telescope’s Space Telescope Imaging Spectrograph (STIS). We examine the connection between nuclear and galactic inflows and outflows by adding long-slit spectra of the host galaxies from Apache Point Observatory. We demonstrate that nearby AGN can be fueled by a variety of mechanisms. We find that the NLR kinematics can often be explained by in situ ionization and radiative acceleration of ambient gas, often in the form of dusty molecular spirals that may be the fueling flow to the AGN.

I discuss feeding and feedback processes observed in the inner few hundred parsecs of nearby active galaxies using integral field spectroscopy at spatial resolutions of a few to tens of parsecs. Signatures of feedback include outflows from the nucleus with velocities ranging from 200 to 1000 km s−1, with mass outflow rates between 0.5 and a few M⊙ yr−1. Signatures of feeding include the observation of gas inflows along nuclear spirals and filaments, with velocities ranging from 50 to 100 km s−1 and mass flow rates from 0.1 to ∼1 M⊙ yr−1. These rates are 2–3 orders of magnitude larger than the mass accretion rate to the supermassive black hole (SMBH). These inflows can thus lead, during less than one activity cycle, to the accumulation of enough gas in the inner few hundred parsecs, to trigger the formation of new stars, leading to the growth of the galaxy bulge. Young to intermediate age stars have indeed been found in circumnuclear rings around a number of Active Galactic Nuclei (AGN). In particular, one of these rings, with radius of ≈ 100 pc is observed in the Seyfert 2 galaxy NGC 1068, and is associated to an off-centered molecular ring, very similar to that observed in the Milky Way (MW). On the basis of an evolutionary scenario in which gas falling into the nuclear region triggers star formation followed by the triggering of nuclear activity, we speculate that, in the case of the MW, molecular gas has already accumulated within the inner ≈ 100 pc to trigger the formation of new stars, as supported by the presence of blue stars close to the galactic center. A possible increase in the star-formation rate in the nuclear region will then be followed, probably tens of millions of years later, by the triggering of nuclear activity in Sgr A*.

Our aim is to study the stellar kinematics in hosts of AGN. We do so by measuring nuclear stellar velocity dispersions (σ⋆). Our sample comprises spectra of 78 objects, 69 of which are Seyfert galaxies, in the region around the λλ8498.02, 8542.09, 8662.14 Calcium triplet (CaT). We investigate two methods to measure σ⋆: (1) the direct fitting method (DFM), which makes use of our stellar population synthesis code Starlight, and (2) the cross-correlation method (CCM), for which we use the fxcor package in IRAF. Both methods yield velocity dispersions consistent to within 19 km/s on-average. We have also measured the CaT equivalent width (WCaT) and the λ3933 CaII K equivalent width (WK) for the objects in our sample. Other studies have shown that WK is a powerful tracer of starbursts in Seyfert nuclei, so we check if WCaT can be used in the same way. We have also analysed a sub-sample of 34 spatially resolved objects with reasonably good off-nuclear signal-to-noise. We study the spatial variations of both σ⋆ and WCaT. We see no dilution in WCaT for composite starburst + Seyfert 2 galaxies, in contrast to the dilution in the WK (studied by other authors) for the same objects.

We present the results of UV spectroscopy of AGNs with double-peaked Balmer emission lines. In 2/3 of the objects, the far-UV resonance lines are strong, with single-peaked profiles resembling those of Seyfert galaxies. The Mg II line is the only UV line with a double-peaked profile. In the remaining objects, the far-UV resonance lines are relatively weak but still single-peaked. The latter group also displays prominent UV absorption lines, indicative of a low-ionization absorber. We interpret the difference in the profiles of the emission lines as resulting from two different regions: a dense, low-ionization accretion disk (the predominant source of the Balmer and Mg II lines), and a lower density, higher-ionization wind (the predominant source of the far-UV resonance lines). These results suggest a way of connecting the double-peaked emitters with the greater AGN population: in double-peaked emitters the accretion rate onto the black hole is low, making the wind feeble and allowing the lines from the underlying disk to shine through. This scenario also implies that in the majority of AGNs, the wind is the source of the broad emission lines.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

We have compared galaxies hosting an active nucleus with non-active galaxies in the SDSS by analyzing their stellar populations. We conclude that the Seyfert 2 phenomenon appears in galaxies of intermediate masses (∼2 × 1010 M$_\odot$), while low mass galaxies do not produce active nuclei, and high mass galaxies tend to produce a low level of non-stellar activity. We also compared the environment of active and non-active galaxies of similar masses and concluded that there is no excess of close neighbors among the Seyferts when compared with non-active galaxies.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

An increasing number of AGNs exhibit broad, double-peaked Balmer emission lines, which arise from the outer regions of the accretion disk which fuels the AGN. The line profiles vary on timescales of 5–10 years. Our group has monitored a set of 20 double-peaked emitters for the past 8 years (longer for some objects). Here we describe a project to characterize the variability patterns of the double-peaked Hα line profiles and compare with those of two simple models: a circular disk with a spiral arm and an elliptical disk.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

We use the photoionization code MAPPINGS Ic to model optical emission-line ratios of the extended emission-line regions of 24 Seyfert galaxies. Although most line ratios can be reproduced by varying the ionization parameter, gas density and/or the shape of the ionizing continuum, the ratio [NII]λλ6548,84/[OII]λ3727 can only be reproduced if we allow for a range of oxygen abundances of 0.5≤ (O/H)/(O/H)$_{\odot} \leq$ 3 for the whole sample. We propose a calibration to obtain (O/H) in terms of [NII]/[OII]. Also, in order to reproduce the emission-line spectra beyond 2 kpc from the nucleus, we had to introduce the contribution from hot stars to the ionizing spectra. We conclude that this is due to an increasing contribution of HII regions with distance from the nucleus to the emission-line regions.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

We announce the construction of an Atlas of STIS-HST Spectra of Seyfert Galaxies, covering the UV and/or optical spectral range. Nuclear spectra of these galaxies have been extracted in windows of width 0.2″ for an optimized sampling (as this is the width of the slit in most cases), from a long-slit spectrum usually covering 52″. Extranuclear spectra have also been extracted, usually using larger windows to improve signal-to-noise ratio. Spectra obtained with different gratings have been combined in order to provide the largest possible wavelength coverage for each galaxy.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

We characterize the stellar population of a complete sample of the closest southern radio-galaxies – 12 Fanaroff-Riley I (FR I), 8 FR II and 4 with undefined FRx type – and a comparison sample of 18 non-active elliptical and SO galaxies. We measured equivalent widths and continuum colours and their radial variations, and performed spectral synthesis as a function of distance from the nucleus. Only 4 of the radio-galaxies have more than 10% of the total flux at 4020Å contributed from 100 Myr old or younger stellar population components – including a featureless continuum due to an AGN. All 4 are FRII or FRx. The main difference between the stellar population of the radio-galaxies and that of the non-active ones is that the former have a larger contribution of the intermediate age component, suggesting a relation between the present radio activity and a past episode of star formation which occurred about 1 Gyr ago.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html

The extended ionized gas in Seyfert and Radio-Galaxies is characterized by large values of the ratio He ii/Hβ, which exceeds the value predicted by the standard photoionization model in which the ionizing continuum consists of a power-law. This has lead to the suggestion of considering a matter-bounded (MB) component for explaining such extreme values. We now find that it is also possible to resolve the temperature problem if the thickness and the ionization parameter of the MB is appropriately selected. Adopting a canonical power law (α= −1.3) and solar abundances (Z=l), we can account for the observed trends in excitation (represented for example by the ratio [0 ii]/[0 iii] in Fig. 1)) by varying the relative number of MB clouds (which emit the high excitation lines Civ, [Ne v], He ii… and most of [O iii]) versus the number of ionization-bounded (IB) clouds (which emit [N ii],[S ii] [O ii], [O i]…). We obtain a one-parameter sequence (solid line) which is function of the weight AM/I of the MB component relative to the IB component. This AM/I–sequence successfully reproduces the observed range in He ii/Hβ. Note the failure of the traditional U–sequence (long dashed line). Fig. 2 indicates that we can also reproduce the ratio ROIII= [O III]λ4363/[O iii]λ5007 and therefore resolve the temperature problem. Interestingly, our model indicates a temperature difference of 5 000 K between the IB component ([N ii] temperature ≃ 10 000 K) and the MB component ([O iii] temperature ≃ 15 000 K) while the traditional U–sequence predicts a difference of only 1 000 K. Such difference of 5 000 K has been reported in the extended gas of Cygnus A.

Observational evidences of dust in the nuclear region of AGNs are substantial (Rudy 1984, ApJ, 284, 33; Jones et al. 1984, PASP, 96, 692). The ionization cones observed in several Seyfert galaxies has been interpreted as shadowing effects by a dust obscuring torus which hides the broad emission line region (BLR) and the central source (Wilson 1992; Storchi-Bergmann, Mulchaey and Wilson 1992, ApJ 395, L73). A large sample of optical and far-IR data for IRAS Seyfert galaxies has been analysed together with dust emission models (Bonatto and Pastoriza 1993), where it has been concluded that the same dust emission model can be applied to both Seyfert types. In order to further study the effects of dust in the spectra of active galactic nuclei, we have obtained spectrophotometry of 21 IRAS Seyfert galaxies in the range 3500–7200 Å and analyse them in conjuction with their IRAS fluxes. The stellar population type is derived from comparisons with normal galaxy templates using dilution effects in the K CaII line as discriminator. For 55% of the sample the population is of late type. For the rest, blue continua due to recent star formation and/or power-law may amount up to 30% at 4000Å. We conclude that the bulge stellar populations of IRAS Seyfert galaxies are similar to those of normal spirals, except that they are more reddened by E(B-V)i ∼ 0.20. Population-subtracted emission line ratios indicate on average stronger reddening for the narrow-line region (E(B-V)l ∼ 0.8. From photoionization models a power-law index for the ionizing continuum α=1.5, and a metallicity larger than solar are obtained. The most luminous IRAS galaxy of the sample (IRAS555) is discuss in detail: in order to be compatible with the observed IRAS fluxes and the optical stellar continuum, the ionizing continuum must be reddened by AV > 10 magnitudes. Consequently a dust structure in this galaxy appears to be increasingly affecting stars and gas towards the galaxy center.

We present the preliminary results of an 8-month monitoring campaign carried out on 6 AGN during the period December 1991 – July 1992. All but one of our targets showed continuum and/or line variability. The data were obtained using the 2D-Prutti + Cassegrain spectrograph at the CTIO 1.0-m telescope, and reduced following standard procedures. The slit width was 5″ and the nuclear spectra were extracted in a 10″ aperture. The wavelength coverage is 3500–7200Å, with 8Å resolution. The data were flux calibrated using standard stars and then normalized using the [O III] λ5007Å line flux for each object. NGC 6814: our spectra reveal that this object is still in a low state of activity and within the S/N ratio of our data, no variability was observed during this campaign. The stellar population is dominant in the nuclear spectrum and a synthesis using the star cluster library of Bica (1988) indicates a mainly old (∼ 86% of the continuum flux at 5870Å due to a population with age ≥ 10 Gyr), [Z/Z⊙] ≥ 0.3 stellar content, with an intrinsic reddening of E(B–V)=0.20. NGC 3227: using an off-nuclear spectrum corresponding to the two 5″×10″ regions 20.4″ E/W of the nucleus, we obtained also a mainly old (77% at 5870Å with age ≥ 10 Gyr), [Z/Z⊙]=0.3) synthetic stellar population, which contributes ∼ 43% of the nuclear light at 5600Å. The cross correlation of the 4245Å continuum and Hβ light curves results in a 18±3 -day lag. IC 4329A: our data show evidence of variability as a slow and constant increase in both continuum and lines fluxes, but no isolated event was detected. ESO141-G55: the light curves show small variations in the continuum, but no noticeable line variability. Akn 120 and Fairall 9: the data consists of two sets of spectra, separated by ∼ 6 months. Within each set little or no variability was detected, but strong line and continuum variations occurred between them.