The evolutionary impact of supermassive black holes on their host galaxies

By exploiting the ultra-deep, Ks-band imaging of the FourStar Galaxy Evolution Survey (ZFOURGE), we investigate the evolutionary impact of supermassive black holes (SMBHs) on their host galaxies. We advance this line of research by applying new analyses to a more substantial number of sources, with higher quality photometric redshifts, over a broader redshift range, and down to deeper mass-limits than most studies before it. We begin by supplementing the ZFOURGE galaxy catalogues with data in radio, X-ray, and infrared wavebands to catalogue AGN host galaxies across a broad redshift range of z = 0:2--3:2. We then use these catalogues to construct a mass-complete, luminosity limited sample of AGN to compare their rest-frame U -- V versus V -- J (UVJ) colours and specific star formation rates (sSFRs) to a mass-matched control sample of inactive (non-AGN) galaxies. Our UVJ diagnostics reveal that the AGN tend to be hosted in a lower fraction of quiescent galaxies and a higher fraction of dusty galaxies than the control sample. Using 160 μm Herschel PACS data, we find the mean sSFRs of AGN hosts to be elevated by 0:34+-0:07 dex with respect to the control sample across all redshifts. This offset is primarily driven by infrared-selected AGN, where the mean sSFR is found to be elevated by as much as a factor of ~5. The remaining population, comprised predominantly of X-ray AGN hosts, is found mostly consistent with inactive galaxies, exhibiting only a marginal elevation. To probe the elevated star formation (SF) found in our AGN samples, we turn our attention to infrared-selected AGN and apply techniques to separate the AGN and SF components of each galaxy’s spectral energy distribution (SED). We use this approach to estimate the black hole accretion rate (BHAR) and star formation rate (SFR) for Milky Way (MW) and Andromeda (M31)-mass progenitors from z = 0:2 -- 2:5. Our motivation here is to minimise the effects of AGN contamination and selection-bias as we track the evolution of the SF-AGN connection of these sources from high to low redshift. Specifically, we track the evolution of their quenching rate via UVJ diagnostics, and their relative black hole-galaxy growth (i.e. their BHAR/SFR ratio). We find as the progenitors evolve, their BHAR/SFR ratio does not track the rate at which progenitors quench. Furthermore, the logarithm of the BHAR/SFR ratio of MW-mass progenitors evolves with a slope of 0:64 -- 0:11, whileM31-mass progenitors are 0:39 -- 0:08. These results contrast with previous studies that find an almost flat slope when adopting X-ray/AGN-selected or mass-limited samples and is likely due to their use of a broad mixture of galaxies with different evolutionary histories. Our use of progenitor-matched samples highlights the potential importance of carefully selecting progenitors when searching for evolutionary relationships between SMBHs and their host galaxies. Finally, we present a new technique to isolate the AGN contribution to the observed SED using only two photometric bands. Such an approach lends itself to studies where abundant photometric data may be lacking. While our approach is not without its limitations, we show it can reproduce underlying trends known to exist in AGN samples. Specifically, we reproduce the SF-AGN correlation found in infrared-selected AGN. However, similar to the previous chapter, we highlight where selection-bias may be driving these results. Overall, this thesis adds significant evidence to the suggestion that correlations observed between SMBHs and their host galaxies may be driven by selection effects, while also casting doubts over the idea that the suppression of star formation is predominantly driven by the negative feedback of luminous AGN over most of cosmic time.