A 3D view of AGN triggering

Résumé

Active galaxies are these showing strong emission coming from their nuclei. In the case of normal galaxies the nuclear emission has a stellar origin, but in active galactic nuclei (AGN) the emission is produced by accretion of matter into a super-massive black hole (SMBH). Although only a small percentage of galaxies in the nearby Universe currently show nuclear activity, it is thought that all massive galaxies might experience episodes of activity for at least some part of their evolution. In fact, AGN are considered as key ingredients to the evolution of galaxies through AGN feedback. Therefore, unveiling the mechanisms that trigger AGN is crucial for our understanding of the formation and evolution of galaxies. While major mergers are often associated with the triggering of powerful AGN, less luminous AGN would be ignited by secular processes, like disc instabilities or bars. Over the last few years, several observational studies performed at optical wavelengths have tried to identify AGN triggering mechanisms in the local Universe. Despite these efforts, the main drivers for AGN triggering in isolated galaxies still remain unclear In this thesis, we try to shed some light on this topic by focusing on isolated nearby spiral galaxies with and without bars. We selected a sample of active galaxies and a control sample of non-active galaxies matched in largescale galaxy properties such as stellar mass, morphology, luminosity, inclination and bar size. Unlike previous works, which have usually identified a sample of active galaxies and compared with the general population of galaxies, we built a sample of pairs of galaxies differing only in nuclear activity (twin galaxies) and performed one to one comparisons. Following this approach, we can identify differences in the kinematics and stellar populations that might be connected to nuclear activity. To do so, spatially resolved observations with spectral information are necessary. We use integral field data from the CALIFA (Calar Alto Legacy Integral Field Area) survey. This survey provides data cubes for ~600 galaxies which are representative of the nearby Universe being a homogeneous set of data (i.e. observed with the same telescope/instrument and reduced with the same pipeline). Firstly, in order to outline the methodology and identify relevant parameters to be analysed using the whole sample, we performed a pilot study. We selected two isolated active galaxies (barred and unbarred) and their corresponding nonactive twin galaxies. We characterised the gas and stellar kinematics, as well as the stellar population properties of these two pairs of galaxies. We performed simple kinematical modelling of the gas and stellar velocity fields considering the whole field, and also the approaching and receding sides. We found that the gas and stellar velocity fields can be modelled by a simple rotating disc at kpcscales. However, the unbarred AGN shows an internal twist in the gas velocity field that might be playing the same role as the bar of the barred AGN in driving an inflow of gas to the central region. We also found that both active galaxies (barred and unbarred) present higher values of the spin parameter (λR) than their corresponding twins. This parameter characterises the rotational support of a galaxy and it is usually used as a proxy of the stellar angular momentum. Additionally, we find that the central regions of the active galaxies show older stellar populations. Secondly, we expanded the study of λR measured in the disc-dominated region to the whole sample of pairs. We found that active galaxies show higher values of λR than their corresponding non-active twins (80% of the pairs). Considering only the unbarred pairs, we found that 100% of the active galaxies show larger λR than their twins. These results indicate that active galaxies present larger rotational support in the disc than non-active galaxies. We suggest that they could be explained by a more efficient angular momentum transfer from the inflowing gas to the disc baryonic matter in the case of the active galaxies. This inflow of gas could have been induced by disc or bar instabilities, although we cannot rule out the effect of minor mergers, which might not be detectable in the shallow SDSS images. Thirdly, we characterised the stellar population properties of the whole sample of pairs. For this, we used full spectral fitting of the optical spectra recovering the star formation history (SFH) of our galaxies. We computed average ages and metallicities, but we also analysed different stellar sub-populations according to their ages (i.e. young, intermediate and old). Additionally, we stacked the spectra of three different regions within the galaxies (bulge-dominated, discdominated and total), and we characterised the chemical enrichment histories of the pairs of galaxies. We found that AGN show a higher tendency to have older stellar populations and are more metal rich than their non-active twins in the bulge-dominated region, indicating that active galaxies have a different chemical enrichment history. Our results suggest that the differences in metallicity would not be predominantly associated with the current active phase but they are prior to it. We rule out the effect of the featureless AGN contribution in the age and metallicity characterisation as the driver of the differences in metallicity found in this work. This thesis provides relevant insights to the emerging picture of the nuclear activity and its effect on the host galaxy. For that, the developed approach (one to one comparisons) has been fundamental. The idea that every massive galaxy goes through a few short active phases during its life is becoming popular. If all galaxies go through active phases, considering that the lifetime of these active phases are estimated to be a tiny fraction of that of galaxies, we should not expect to find large-scale differences between the twins either in dynamics or in stellar population properties over longer-timescales than the current AGN episode. Different active phases in both the active and non-active twins should dilute them. However, the results presented here support the idea that some galaxies are more likely to go through active phases than others, at least in the redshift (0.005 < z < 0.03) and mass ranges considered in this thesis.