Incidence and energetics of AGN winds in the distant Universe

The initial growth phases of active galactic nuclei (AGN) and their central supermassive black holes (SMBHs) are crucial in galaxy evolution, with AGN feedback playing a significant role. However, selecting sources in the feedback phase remains challenging. This thesis focuses on the selection and characterization of AGN in the feedback phase, emphasizing the study of ionized outflows and their impact on host galaxies. The first part presents methods used to isolate ~1400 AGN candidates from the eROSITA Final Equatorial Depth Survey (eFEDS) catalog and the outflow properties derived for a subsample of 23 sources with confirmed ionized winds. For z > 0.5 AGN with detected ionized outflows, we report correlations between mass outflow rate, outflow kinetic power, and AGN bolometric luminosity. However, a weak correlation between maximum outflow velocity and AGN bolometric luminosity highlights the importance of sample selection. Spatially resolved spectroscopic data from MUSE was used to study the morphology and kinematics of ionized gas in a red, obscured quasar within an interacting system. Extended ionized outflows with high velocities and mass outflow rates were observed, likely AGN-driven rather than star formation-driven. However, kinetic coupling efficiencies were low, suggesting the outflows are not energetically significant. Finally, machine learning techniques enhanced AGN selection in the feedback phase. A Random Forest Classifier achieved 95% accuracy in classifying AGNs into feedback and non-feedback phases. This work demonstrates the potential of machine learning in identifying AGN feedback phases in large-scale surveys like eRASS1. This thesis provides insights into the selection, characterization, and impact of AGN outflows on their host galaxies, advancing our understanding of galaxy evolution. Findings suggest that outflows may or may not significantly impact host galaxies, depending on kinetic coupling efficiencies.