Observational constraints on the influence of active galactic nuclei on the evolution of galaxies

Abstract

At the centre of every massive galaxy there resides a super-massive black hole that grew during periods of active galactic nuclei (AGN) activity. Current theoretical models of galaxy evolution invoke AGN-driven galaxy-scale feedback processes (e.g., the expulsion of gas through outflows) in order to reproduce many of the fundamental properties of galaxies and the intergalactic medium. This thesis uses observations to test some of the predictions of these feedback
processes. I present integral field unit observations of AGN host galaxies to trace their ionised gas kinematics. The targets are eight high-redshift (z=1.4-3.4) ultra-luminous infrared galaxies, that are representative of rapidly evolving distant galaxies, and sixteen z<0.2 luminous AGN, that were selected from a parent sample of ~24,000 sources to be representative of the overall population. In both samples I identify galaxy-wide outflows (i.e., over kiloparsec scales), calculate their properties (e.g., mass outflow rates and energetics) and show that they are broadly consistent with theoretical predictions. I find that ionised outflows are common in z<0.2 luminous AGN and are consistently extended over kiloparsec scales (in >70% of cases). I also use far-infrared Herschel data of X-ray detected z=1-3 AGN to test the prediction that luminous AGN shut down star formation in their host galaxies. Using stacking techniques I show that, on average, X-ray detected AGN over a wide range of luminosities (i.e., L(2-8keV)~10^42-10^45) erg/s) have star formation rates that are consistent with non-active galaxies. This may imply that luminous AGN do not impact upon the star formation in their host galaxies. Overall, this thesis demonstrates that energetic galaxy-wide outflows are prevalent in AGN host galaxies; however, it also demonstrates that we still lack direct observational evidence that luminous AGN are suppressing star formation in their host galaxies.