The RASS-SDSS galaxy cluster survey.

Galaxy clusters are the largest gravitationally bound systems in
the universe. Clusters consist of three components: galaxies, gas,
and dark matter. The galaxies themselves contribute the least, at
most a few percent, to the total mass. The remainder consists of
diffuse, hot gas (the intracluster medium, or ICM) and an unseen
component which is needed to explain the gravitational stability of
clusters (the dark matter). The two most obvious means of studying
clusters of galaxies are by observing the optical light emitted
from the constituent galaxies or the X-ray emission from the ICM.
Clusters of galaxies, bound ensambles of hundreds of galaxies, are
an ideal environment to study galaxy evolution and to learn how
this is affected by different physical processes: gravity,
starbursts and star formation, interactions with the intergalactic
medium and galaxy-galaxy encounters. Since the very early works of
Hubble in the thirties, it has been recognized that galaxies in
dense environments differ systematically from those in low-density
regions in their morphological types, stellar populations and
gaseous content. When during the history of the Universe and why
such environmental differences were established is currently one of
the subjects of most intensive investigation in the international
astrophysical community. On the other hand, clusters can teach us a
great deal about cosmology. The distributions of galaxies on the
sky shows a net-like structure in which thin walls and filaments
surround large voids. The galaxy clusters are the nodes of this
network. Therefore, they trace out the Large-Scale Structure (LSS)
of the universe and can be used to study the LSS formation.
Moreover, if clusters provide a 'fair sample' of the universe, then
the fraction of their mass in baryons should equal the universal
baryon fraction, known as $\Omega_b/\Omega_m$. Moreover, the
evolution of cluster number density with redshift can determine the
mass density parameter, known as $\Omega_M$, and possibly determine
the equation of state (and nature) of the dark energy believed to
be causing the expansion of the universe to accelerate. Thus,
galaxy clusters have a twofold importance: first as laboratories of
galaxy formation and evolution, and second as cosmological tool.
The aim of this project is to study galaxy clusters from these two
perspectives. For this purpose we use the largest optical and X-ray
surveys ever realized, the Sloan Digital Sky Survey (SDSS) and the
Rosat All Sky Survey (RASS), respectively, to conduct a
multiwavelenght study of the properties of galaxy clusters. The
project is called RASS-SDSS Galaxy Cluster Survey reflecting the
name of the two big surveys used for this work. All the analyses
are performed on two cluster samples specially created for the
survey: the X-ray selected RASS-SDSS galaxy cluster catalog and a
subsample of optically selected, isolated and spectroscopically
confirmed Abell clusters. The project consists of two parts. The
aim of the first part is to understand which role play the
gravitational processes, galaxy mergers and collisions and the
interaction with ICM in the process of galaxy formation and
evolution. For this purpose, we study the variations of several
properties of the cluster galaxy population such as the luminosity
and spatial distribution, the morphological type mix, the Star
Formation Rate (SFR) and stellar mass as a function of the
environmental conditions and the cluster global properties. Our
detailed analysis of the cluster individual and composite
luminosity functions reveals that the LF clearly shows a bimodal
behavior with an upturn and a evident steepening in the faint
magnitude range in any SDSS band. The LF is well fitted by the sum
of two Schechter functions. The bright end of the LF is found to be
universal in all the clusters. The faint end of the LF is much
steeper and varies significantly from system to system, when
calculated within a fixed metric aperture. The variations are not
ramdom however. The more massive a cluster, the lower its fraction
of dwarf galaxies. This effect disappears when the cluster LF is
calculated within the physical size of the system, as the virial
radius ($r_{200}$). This indicates that the previously observed
variations are due to aperture effects caused by the observed
increase of the fraction of dwarf galaxies with the clustercentric
distance. Our conclusion is that the shape of the cluster LF is
universal in all the magnitude ranges when the LF is calculated
within the virial region. Moreover, the analysis of the composite
cluster LF per morphological type, shows that the upturn and the
steepening at the faint end of the LF is caused by dwarf early type
galaxies. These systems are quite rare in low density regions and
appear to be a typical cluster population. We provide evidence that
the process responsible for creating the excess population of dwarf
early type galaxies in clusters is a threshold process that occurs
when the density exceeds $\sim 500$ times the critical density of
the Universe. We interpret our results in the context of the
'harassment' scenario, where faint early-type cluster galaxies are
predicted to be the descendants of tidally-stripped late-type
galaxies. In the same context, we investigate whether the cluster
total star formation rate ($\Sigma SFR$) depends on the cluster
global properties for a sample of 90 very nearby clusters. The
total cluster SFR is given by the sum of the SFR of all the cluster
members within the virial region. It is found to be proportional to
the number of cluster galaxies involved ($N_{gal}$). The best
relation between the total SFR and the cluster mass reflects the
$N_{gal}-M$ relation, which is a power law with exponent smaller
than 1. As a consequence, the more massive a cluster, the lower its
number of cluster galaxies and total SFR per unit mass. The mean
SFR per cluster galaxy ($\Sigma SFR/N_{gal}$) is constant troughout
our cluster sample and does not depend on the global properties of
the system. Moreover, in order to account for projection effects,
we study the galaxy surface number density profile in our cluster
sample. We find that clusters of different mass exibit different
profiles. In the low and intermediate mass systems the best fit is
provided by a core King profile, with the core radius decreasing
with cluster mass, until, at the highest cluster masses, the
profile is better represented by a cuspy Navarro, Frenk \&
White profile. All these different analysis converge to the
conclusion that the global properties of the cluster galaxy
population, such as the luminosity distribution, the galaxy type
mix, the mean and total cluster SFR are only weakly dependent on
the cluster mass and X-ray luminosity. This suggests that the
gravitational processes and the interaction galaxy-ICM are not
likely to affect those properties of the cluster galaxy population.
Only the spatial distribution of the cluster galaxies depends on
the cluster mass, probably reflecting the different relaxation
status of systems of different masses. Instead, the variations of
the LF and the galaxy type mix with the clustercentric distance
reflect a link between the galaxy formation process and the
galaxy-galaxy encounters, as suggested by the 'harassment'
scenario. In the second part of the thesis, galaxy clusters are
used as cosmological tool. The aim of this work is to elucidate
which component, galaxies or ICM, traces better the cluster mass in
order to understand whether different selection methods select the
same cluster population. This will clarify which bias is introduced
by the different selection methods in the results of the
cosmological tests. This will clarify which bias is introduced by
the different selection methods in the results of the cosmological
tests. For this porpuse, we analyse as a first step the relation
between the optical ($L_{op}$) and the X-ray ($L_X$) luminosity,
respectively, to the cluster mass in the X-ray selected RASS-SDSS
cluster sample. The main motivation in deriving these dependences
is to evaluate $L_{op}$ and $L_X$, as predictors of the cluster
mass and to compare the quality of the two quantities as
predictors. Our analysis reveals that $L_{op}$ is a key measure of
the cluster mass. In this respect, the optical luminosity performs
even better than the X-ray luminosity, which suggests that the mass
distribution of a cluster is better traced by cluster galaxies
rather than by intracluster gas. On the other hand, our conclusion
is at odds with the generally accepted view that a cluster main
physical properties are more easily revealed in the X-ray than in
the optical. Such a view was established at an epoch when the lack
of optical wide field surveys precluded a reliable determination of
the optical luminosities of a large sample of clusters. With the
advent of the Sloan Digital Sky survey, this problem is now
overcome. The application of the same analysis to an optically
selected cluster sample (the Abell subsample) confirms the result.
Neverthless, the Abell sample comprises a subpopulation of systems
which scatter significantly in the $L_X-M$ relation and appear to
be extremely X-ray underluminous (on average one order of
magnitude) with regard to their mass. On the other hand, these
systems do follow the general scaling relation between optical
luminosity and virial mass. Therefore, we call them 'Abell X-ray
Underluminous clusters' or AXU clusters for short. To understand
the particular nature of these systems, we examine the properties
of their galaxy population. The velocity distribution of the AXU
clusters is Gaussian within the virial region but is leptokurtic
(more centrally concentrated than a Gaussian) in the outskirts, as
expected for the systems in accretion. In addition, the AXU
clusters have a higher fraction of blue galaxies in the external
region and show a marginally significant paucity of galaxies at the
center. Our results seem to support the interpretation suggested by
Bower et al. (1997) that the AXU clusters are systems in formation
undergoing a phase of mass accretion. Their low X-ray luminosity
should be due to the still accreting Intracluster gas or to an
ongoing merging process. Our results give supports to the
conclusion of Donahue et al. (2002) concerning the biases inherent
in the selection of galaxy clusters in different wavebands. While
the optical selection is prone to substantial projection effects,
also the X-ray selection is not perfect or not simple to
characterize. The existence of X-ray underluminous clusters, even
with large masses, makes it difficult to reach the needed
completeness in mass for cosmological studies. Clearly, a
multi-waveband approach is needed for optimizing the completeness
and reliability of clusters samples. The 'RASS-SDSS Galaxy Clusters
Survey' series comprises 7 scientific papers which are inserted as
part of the thesis. Four of the papers are accepted for
pubblication on a scientific Journal ('Astronomy &
Astrophysics') and three are submitted.