Large-scale structure studies using AGN in X-ray surveys

Large X-ray surveys are a powerful tool to study the large-scale
structure (LSS) of the Universe. The scientific impact of LSS
studies using active galactic nuclei (AGN) in X-ray surveys can be
significantly increased by conducting wider and deeper X-ray
surveys and studying the surface brightness fluctuations of the
unresolved cosmic X-ray background (CXB). In the first part of this
Thesis, we have investigated the prospects of using the AGN sample
to be detected by the upcoming eROSITA all-sky survey (eRASS) for
LSS studies. We show that eRASS will detect about 3 million AGN in
the 0.5-2.0 keV band. This will result in a ~30 times larger number
of sources and a ~30 times better sensitivity than its 25 year old
predecessor, the ROSAT all-sky survey (RASS). We show that this
unprecedented AGN sample will have a median luminosity of ~10^44
erg/s, which is typical for the entire AGN population in this
energy band. It will have a median redshift of z ~ 1 and
approximately 40% of the objects will be in the redshift range of z
= 1-2, where the bulk of the X-ray emission of AGN is produced.
About 10^4 - 10^5 AGN are predicted to be beyond redshift z = 3 and
about 2 000 - 30 000 beyond z = 4, which will potentially include
some of the earliest AGN in the Universe. We demonstrate that,
given these unique properties, the eRASS-AGN sample will be able to
significantly improve our current knowledge of the AGN spatial
density as a function of redshift and luminosity over a wide range
of cosmic time. Further, we show that it will enable us, for the
first time, to perform detailed redshift- and luminosity-resolved
studies of the clustering strength of X-ray selected AGN. All these
measurements will dramatically improve our understanding of the
growth of supermassive black holes over cosmic time and its
implications for galaxy evolution. We demonstrate for the first
time that, given the breadth and depth of eRASS, it will be
possible to use AGN as a cosmological probe via baryon acoustic
oscillation (BAO) measurements. We will be able to convincingly
detect BAOs in the currently uncharted redshift range of z ~ 1-2,
which will improve the constraints on the current cosmological
model. In the second part of this Thesis, we have conducted the
most accurate measurement to date of the brightness fluctuations of
the unresolved CXB in the 0.5-2.0 keV band for angular scales of
< ~17'. For this we used the XBOOTES survey, the currently
largest continuous survey of the X-ray telescope Chandra. We find
that on small angular scales (< ~2') the observed power spectrum
of the brightness fluctuations is broadly consistent with the
conventional AGN clustering model, although with a 30% deviation.
This deviation nevertheless presents a good opportunity to improve
our understanding of clustering properties of unresolved AGN by
testing more sophisticated clustering models with our measurement.
For angular scales of > ~2' we measure a significant excess with
up to an order of magnitude difference in comparison to the
standard AGN clustering model. We demonstrate that an instrumental
origin can be excluded. However, we also show that the excess can
neither be explained with any known X-ray source population by
looking at strength of its clustering signal and the shape of its
energy spectrum. It might be caused by more than one type of source
but the dominant source appears to have extragalactic origin.
Finally, we make predictions on how eRASS will be able to advance
the studies of the unresolved CXB.