AGN outflows
Astrophysical outflows can be found in a large number of diverse environments and have been associated with different processes that span several orders of magnitude in spatial and temporal scale and energetics. Beyond those associated with stellar processes, some of the most energetic outflows are linked to supermassive black holes (SMBHs), ejected during phases of active accretion and are therefore associated to active galactic nuclei (AGN). Such outflows have been observed for all the different phases of the interstellar medium (ISM) gas: hot plasma (radio jets from closest to the SMBH), ionized gas (launched from the accretion disk around the SMBH and reaching well into the host galaxy, and molecular gas.
AGN outflows & feedback
Luminosity function of galaxies in the local Universe (blue points) compared with cosmological simulations without and with AGN feedback (dashed and solid black lines, respectively). The most luminous (and most massive) galaxies are overproduced in the simulations if AGN feedback is not considered. This means the host galaxy growth needs to be regulated in some way.
AGN-driven outflows are considered to be manifestations of the AGN feedback, which theoretically also seems to be necessary in order to accurately reproduce our present day Universe in cosmological simulations. Galaxy-wide outflows, now understood to be prevalent in luminous AGN, are thought to be an efficient way of coupling the AGN energy output to the galactic interstellar medium (ISM). These outflows accelerate gas to high velocities and presumably eject it from the gravitational potential of its host. Correlations between the outflows size and kinetic energy and AGN luminosity imply that they are radiation-driven.
Warm gas outflows driven by AGN
We have identified the best candidates for Type 2 AGN with outflows in the local Universe. Our selection is based on the SDSS and is based on the [OIII] luminosity of the AGN and the kinematic properties of the [OIII] line from the SDSS spectrum. Particularly, we select sources with log-luminosity about 42 erg/s (after correcting for dust extinction) and absolute projected velocities larger than 250 km/s (w.r.t. systemic velocity of the stars) or velocity dispersion larger than 400 km/s (or roughly ~2 times the stellar velocity dispersion for these galaxies).
These criteria select approximately 0.5% of all Type 2 AGN (29 objects) in the local Universe, making them a truly unique sample of the best sources to study outflows driven by AGN. We followed these up with integral field unit (IFU) spectroscopy to get spatially resolved in formation about the line emitting warm gas. For that we used telescopes around the world, including the Gemini-North and UH2.2m telescopes in Hawaii and Magellan in Chile.
Outflow kinematics
Our observations allow us to look at the spatially resolved kinematic properties of these outflows and study how they change as a function of distance from the AGN, but also with respect to the luminosity of the AGN or other characteristic properties of the system.
We find that the [OIII] emitting gas shows extreme kinematics, with velocities of up to 800 km/s at the center of the galaxy and similarly large velocity dispersions. The region affected by these extreme kinematics however appears rather compact, and in some cases is below the resolving power of our data (we are using seeing-limited observations with seeing of ~0.5-0.8 arcsecs).
The Ha emitting gas on the other hand shows somewhat different kinematic patterns, with strong signs that a significant fraction of the Ha-emitting gas is actually following the stellar rotation within the gravitational potential of the galaxy. Given our stellar light and emission line fitting, we are able to decompose these different kinematic components and study them separately.
Related papers: Karouzos et al. 2015b (in prep.)
AGN photoionization and circumnuclear star formation
Given our wide wavelength coverage, we are able to use the different emission lines to classify the mechanism that is ionizing the gas within different regions of the galaxy. We find, as expected, that the central part of all our sources and photoionized by the AGN. However, at the same time we also recover significany evidence for both shock-induced ionization and photoionization from young stars, which complicate the picture!
We do a kinematic decomposition of the spatially-resolved emission line flux ratios to associate line-emitting gas at different velocities with the different ionization mechanisms. We find that the most extreme kinematics (blueshifted emission) is exclusively associated with photoionization from the AGN. In contrast, redshift emission and emission around the systemic velocity have a strong contribution from ongoign star formation.
Related papers: Karouzos et al. 2015c (in prep.)
Future plans
We have now concluded the follow-up IFU observations for all of our targets and are in the process of analyzing the data. The analysis of the Gemini/GMOS data of 6 targets is concluded and we will be submitting two papers in the following months. We have also follow-up observations using long-slit spectroscopy for radio-loud AGN, to investigate potential jet-driven outflows. A particularly intriguing aspect of our results concerns the detection of circumnuclear star formation for most of our outflow AGN. The questions then is whether there is a causal connection between the two. We have proposed for further Gemini observations to investigate this connection in detail.