Evidence for Dark Matter from the Dynamical

Mass of Galaxy Clusters

Project report submitted in partial fulfilment of the requirement for the
award of the degree of Master of Science in Physics

University of Kerala
by
Julie Jacob Thomas
Reg. No. 1118106
[Link]. Physics
Under the Guidance of

Dr. Anand Narayanan

Associate Professor
Department of Earth and Space Science
Indian Institute of Space Science and Technology
Valiamala, Kerala – 695547

MAR IVANIOS COLLEGE (AUTONOMOUS)

Department of Physics

Mar Ivanios College Road, Bethany Hills, Nalanchira P.O

Thiruvanathapuram, Kerala 695015
 Preface

The advancement of civilization paved the way for humans to lift their eyes from the
ground and look up at the sky to ponder fundamental questions about the natural
world. Scientists throughout ancient and modern history have devised novel methods
and sophisticated instruments to probe the universe at the microscopic, macroscopic
as well as astronomical scales to find answers to these questions. The fields of
astrophysics and cosmology were born out of the innate curiosity and desire of human
beings to know the secrets underlying the cosmos.

Periodically, every area of science makes a breakthrough through a major discovery

that transforms the existing paradigm of understanding. Further investigations may
lead the scientific community to leave behind old frameworks in favour of new
theories and models that emerge from such a breakthrough. Other times, it may leave
scientists with more questions than before, opening up new avenues of research for
future generations. One such instance that has been gaining momentum in the field of
cosmology in recent years is the discovery of dark matter.

Studies from the previous century, by scientists such as Fritz Zwicky, revealed that
the dynamics of the universe is governed not by the gravity of the stars and galaxies
but by that of matter lying within and around the clusters that cannot be seen or
detected directly, called dark matter (as well as another mysterious entity called dark
energy). The massive improvements in our ability to collect precise data as well as to
test theories and models more rigorously have made it possible to confirm this
discovery in later years; however, despite strenuous and sustained effort from the
astrophysicists and cosmologists, we are yet to arrive at a definitive model that
explains what the nature of dark matter is and how it affects the evolution of the
structure of the universe. Rather, new cosmological models emerge regularly that
challenge or substantiate existing ones and deepen and advance our current
understanding of the large scale structure of the universe.

In this project, the evidence for the presence of dark matter is derived by applying the
standard virial theorem to galaxy clusters, based on F. Zwicky‘s original work
published in 1933. The work involved estimating the mass of a few known
galaxy clusters from galaxy spectroscopic data from the Sloan Digital Sky
Survey, using the kinematics of galaxies that are members of the cluster, and
inferring the presence of dark matter.

Chapter 1 gives an introduction to galaxy clusters and dark matter, and briefly
discusses the evidence found for dark matter within clusters as well as in other
contexts. The review of literature section includes some of the early papers that used
the virial theorem method of estimating cluster masses for the first time as well as
some works from later years that are relevant to this project. It also includes a brief
introduction to the method of calculating dynamical mass adopted in this project, its
limitations as well as the future scope of this work. It also contains some papers that
are focused on mass estimation of galaxy clusters through other methods. Chapter 2 is
a detailed description of the application of virial theorem to the clusters to determine
their dynamical mass. Chapter 3 contains details about the sources of the data, as well
as the data analysis tools, programmes and techniques used in the project. Finally,
Chapter 4 outlines the results of the analysis performed on the galaxy clusters, along
with the conclusions drawn from these results.
 Contents

1 DARK MATTER IN GALAXY CLUSTERS 1-8

1.1 Introduction 1
1.2 Galaxy Clusters 1
1.2.1 Mass-to-light Ratio 3
1.2.2 Crossing Time Scale 4
1.3 Dark Matter 4
1.3.1 The ΛCDM Model 5
1.3.2 Evidence for Dark Matter in Galaxy Clusters 6
1.4 Aim of the Study 7
1.5 Scope of the Study 7
1.6 Limitations of the Study 8

2 REVIEW OF LITERATURE 9-10

3 DETERMINING DYNAMICAL MASS OF GALAXY CLUSTERS 11-15

3.1 Introduction 11
3.2 Virial Theorem Method 11
3.2.1 Characteristic Velocity Dispersion 12
3.2.2 Estimating the Radius of the Cluster 13
3.2.3 Law of Cosines in Spherical Coordinates 14
3.2.4 Angular Diameter Distance 15

4 DATA ANALYSIS 17-25

4.1 Introduction 17
4.2 Sloan Digital Sky Survey (SDSS) 17
4.2.1 SDSS Data Release 7 19
4.2.2 GMBCG Cluster Catalog 19
4.3 Python 20
4.3.1 SciPy 20
 4.3.2 CSV Module 21
4.4 Data Analysis 21
4.4.1 Data Analysis of Cluster GMBCG J197.87292-01.34109 22
4.4.2 Results for Cluster GMBCG J197.87292-01.34109 23
4.4.3 Data Analysis of Cluster GMBCG J231.03093+29.88897 24
4.4.4 Results for Cluster GMBCG J231.03093+29.88897 25
4.4.5 Data Analysis of Cluster GMBCG J47.12599 -0.353214 26
4.4.6 Results for Cluster GMBCG J47.12599 -0.353214 27
4.5 Conclusions 28

A APPENDIX 29-40
A.1 Python Code for Cluster GMBCG J197.87292-01.34109 29
A.2 Python Code for Cluster GMBCG J231.03093+29.88897 33
A.3 Python Code for Cluster GMBCG J47.12599 -0.353214 37

BIBLIOGRAPHY 41-45
 List of Figures

1.1 False-colour mosaic of the central region of the Coma Cluster made with 2
data from the Sloan Digital Sky and NASA's Spitzer Space Telescope [2]

3.1 Spherical triangle solved by the law of cosines. 14

3.2 Definition of equatorial coordinates system on the sky. Observer is at the 15

center of the sphere, which has an arbitrary radius of ∞

4.1 An SDSS Image of Galaxy Cluster GMBCG J197.87292-01.34109 with 22

(RA, Dec) = (197.87292, -01.34109) and redshift z = 0.203.

4.2 Histogram showing distribution of angular separations for all possible 23

pairs of galaxies in the Cluster GMBCG J197.87292-01.34109. The size
of the cluster was chosen as 0.35 after eliminating extreme values.

4.3 An SDSS Image of Galaxy Cluster GMBCG J231.03093+29.88897 with 24

(RA, Dec) = (231.03093, 29.88897) and redshift z = 0.114.

4.4 Histogram showing distribution of angular separations for all possible 25

pairs of galaxies in the Cluster GMBCG J231.03093+29.88897. The size
of the cluster was chosen as 0.4 after eliminating extreme values.

4.5 An SDSS Image of Galaxy Cluster GMBCG J47.12599-0.353214 with 26

(RA, Dec) = (47.12599, -0.353214) and redshift z = 0.374.

4.6 Histogram showing distribution of angular separations for all possible 27

pairs of galaxies in the Cluster GMBCG J47.12599-0.353214. The size of
the cluster was chosen as 0.45 after eliminating extreme values.
 Chapter 1

DARK MATTER IN GALAXY CLUSTERS

1.1 Introduction

It is now known that only about 10% of the mass in the luminous or visible parts of
the galaxy clusters can account for the total mass of the clusters. The rest has been
found to be dominated by dark matter. Zwicky studied the Coma Cluster in the early
1930s and was the first to propose the existence of dark matter within clusters with
the backing of data. Further studies have been carried out on numerous other clusters
through various new methods and approaches and all of them reveal the same result—
a huge discrepancy in the visible mass when compared with the visible or observed
mass. Galaxy clusters are now one of the best probes available to cosmologists and
astrophysicists to study dark matter and large scale structure formation in the
universe.

1.2 Galaxy Clusters

Galaxies seldom exist as lone entities. Galaxies that exist together in a small region of
space decouple from the expansion of space-time due to their enormous gravity to
form groups of galaxies or clusters of galaxies (when there are a large number of
galaxies) or even clusters of clusters called superclusters. Galaxy clusters are
structures that contain up to thousands of galaxies within a relatively small volume.
They are bound, virialized, high overdensity systems of galaxies held together by the
cluster‘s self-gravity [1]. The Milky Way galaxy exists in the galaxy group called the
Local Group which contains galaxies such as the Andromeda Galaxy, the Triangulum
Galaxy, the M31 galaxy, NGC 6822 and smaller satellite galaxies M32 and M110,
among others. The Virgo Cluster and Fornax Cluster are two clusters that lie closest
to the Earth. The Coma Cluster and Perseus clusters are two other prominent clusters.
Figure 1.1 is a false-colour image of the Coma Cluster.

1
 Figure 1.1 This false-color mosaic of the central region of the Coma Cluster combines infrared and
visible-light images to reveal thousands of faint objects (green), many of which are dwarf galaxies
belonging to the cluster. Two large elliptical galaxies, NGC 4889 and NGC 4874, dominate the
cluster's center. The mosaic combines visible-light data from the Sloan Digital Sky Survey (blue) with
long- and short-wavelength infrared views (red and green, respectively) from NASA's Spitzer Space
Telescope [2].

Clusters that contain less than a hundred galaxies are called poor clusters, whereas
clusters with more than hundred galaxies are called rich clusters. They are the largest
stable systems in the known universe, having diameters between 2 and 10 Mpc.
Clusters usually contain elliptical and S0 type galaxies along with several dwarf
galaxies. Typically, their (dynamical) mass is in the range of 1014 to 1015 and
luminosity is in the range 1011 to 1012 .

Clusters contain gas at around 108 K and are very luminous sources of x-rays [3].
They contain a unique mixture of galaxy types, including the biggest and many of the
oldest. The spread of the velocities of individual galaxies within the cluster can range
from a few hundred to a few thousand km/s.

2
They are made of both baryonic (containing neutrons and protons) and non-baryonic
matter. They are mainly composed of three components:

 Galaxies (baryonic), constituting up to 5% [1] of the cluster‘s total mass;

 Intra cluster medium (ICM) containing hot gas and plasma (baryonic),
constituting up 10% of the total mass; and
 Dark matter (non-baryonic), constituting about 85% of the total cluster
mass.

Clusters of galaxies are the most massive virialized systems known and provide
powerful tools in the study of cosmology and extragalactic astronomy. Because of
their large masses and detectability over a large fraction of the age of the universe,
clusters of galaxies provide test ground for cosmology [4]; for example, clusters are
efficient tracers of the large-scale structure in the universe and aid in determining the
amount of dark matter on Mega parsec scales [4].

In the past decade, great advances have been made in our understanding of clusters of
galaxies, from the standpoint of their internal structure and evolution to their place in
the larger scale structure of the Universe. These advances have been made possible by
tremendous improvements in theoretical modeling and numerical simulation, as well
as a wealth of new information provided by multi-wavelength surveys of the Universe
[5].

1.2.1 Mass-to-light Ratio

A conventional method of quantifying the comparison between dynamical mass and

luminous mass is through the mass-to-light ratio of the body. The mass-to-light ratio
is the ratio of total mass of a spatial volume, typically containing a galaxy or a cluster
of galaxies, and its luminosity. The mass-to-light ratio is expressed in units of the
mass-to-light ratio of the Sun, which is a constant equal to 5133 kg/W or equal to the
solar mass divided by the solar luminosity , i.e., .

Any galaxy that is composed solely of stars like our Sun (intermediate and low mass
stars) would have a mass-to-light ratio approximately equal to unity but it can go up
to 10 . However, dynamical mass estimations of several clusters have yielded
values of mass-to-light ratio that is close to 200 to 300 times the mass-to-light ratio.

3
1.2.2 Crossing Time Scale

The galaxy cluster is a gravitationally bound entity. The velocities of individual

galaxies are governed by cluster dynamics—just like an individual, who could move
about freely before entering a crowd, loses several degrees of freedom after entering
it. On entering the crowd, over time, his movement becomes restricted and governed
by the overall movement of the crowd, rather than being able to move freely. The
same logic is applied to clusters.

The cluster is an amorphous structure since it does not have a well-defined boundary.
Let us assume that the characteristic radius of the cluster defines the boundary of the
cluster. The time taken by a galaxy to cross from one end of the cluster to the opposite
end passing through the centre of the cluster is called the crossing time tcross and is
given by the characteristic velocity divided by the characteristic radius of the cluster.
If it is found to be sufficiently small when compared with the age of the universe, it
can be inferred that since tcross is much lesser than the age of the universe, it is safe to
assume that the member galaxies in clusters have had enough time to enter a stable
state within the cluster in which its dynamics are primarily governed by that of the
cluster.

1.3 Dark Matter

The two dominant components of the universe appear to be a non-baryonic form of

matter called dark matter, whose gravity is responsible for structure formation, and a
mysterious form of dark energy, whose pressure is currently causing the expansion of
the universe to accelerate [6].

Dark matter cannot be detected directly since it does not interact with ordinary or
baryonic matter or with electromagnetic radiation, except through gravity. In fact, its
presence has been detected only through its gravitational effects on the dynamics of
structures like galaxy clusters, which cannot otherwise be explained with the currently
accepted theories of gravitation. Currently, the natures of dark matter as well as dark
energy remain unknown.

Dark matter makes up about 85% of the total amount of matter in the universe and is
responsible for large-scale structure formation by drawing the baryonic gas and other

4
matter into deep potential wells, breaking off from the expansion of space and
allowing a part of it to form stars, galaxies and clusters of galaxies. Dark matter
initiates the process of galaxy formation, but, once stars begin to form, supernova
explosions and disturbances wrought by supermassive black holes can inhibit further
star formation by pumping thermal energy into the universe‘s baryonic gas [6].

The concept of dark matter was first hypothesized by Lord Kelvin in 1884 when he
estimated the number of ―dark bodies‖ in the Milky Way galaxy from measuring the
velocity dispersions of the stars that orbited around the Milky Way‘s centre. He
concluded from these measurements that the mass of the Milky Way was different
from what was estimated from the visible stars, and that the sources of this missing
mass were dark bodies. Later, the Dutch astronomer Jacobus Kapteyn, in 1922, and
fellow Dutchman and radio astronomy pioneer Jan Oort, in 1932, also hypothesized
the existence of dark matter.

Dark matter is classified into three types according to their velocity:

 Cold dark matter

 Warm dark matter
 Hot dark matter

Currently, cosmological models favour a cold dark matter universe in which

structures form by the gradual accumulation of smaller bodies.

1.3.1 The ΛCDM Model

The ΛCDM or Lambda Cold Dark Matter Model is a cosmological model in which
the universe consists of three primary components: a cosmological constant Λ, or
Lambda which is associated with dark energy and the accelerated expansion of the
universe, the CDM or cold dark matter and ordinary matter. This model is also
commonly referred to as the Standard Model in Big Bang cosmology. This is because
it gives a relatively simple explanation for phenomena such as the Cosmic Microwave
Background (CMB), the large-scale distribution of galaxies and clusters of galaxies in
the universe, the abundance of hydrogen, and the accelerated expansion of the
universe [7].

5
The model is based on the General Theory of Relativity proposed by Einstein. The
equations of state in this model are governed by the cosmological parameters which
add up to give the total density parameter , which at value is the condition for
critical density. Beyond the critical density, the universe would continue to expand
forever. A few of the cosmological parameters are explained for the purpose of this
project: The cosmological parameter or dark energy density parameter is the
fraction of the flat universe that is dark energy. The parameter is the mass
density or the fraction of the ordinary plus dark matter in the universe. Another
important parameter is the Hubble parameter or Hubble‘s constant which is taken
as 69.6 km/s/Mpc in this project.

1.3.2 Evidence for Dark Matter in Galaxy Clusters

Primary evidence for dark matter comes from calculations showing that many
galaxies would fly apart, or that they would not have formed or would not move as
they do, if they did not contain a large amount of unseen matter [8].

The first such evidence was reported in 1933 by F. Zwicky who estimated the mass of
the Coma Cluster from the velocity dispersions of its member galaxies. He was able
to show that the velocity dispersions and, subsequently, the mass measured by this
method were far too great to be accounted for with the visible or luminous mass
which is estimated based on brightness and number of galaxies within the cluster.
Moreover, the velocities of the individual galaxies were too high for it to be held by
the gravity of the visible mass alone. He concluded that the cluster was dominated by
some form of unseen matter that held the galaxies together in equilibrium, which he
called dunkle Materi—German for dark matter.

Further studies on other clusters using improved methods of mass estimation

increased the estimates of luminous mass, but still could not account for the vast
majority of the unseen matter that dominated the cluster dynamics. Gravitational
lensing observed around galaxy clusters have contributed to our understanding of dark
matter by providing a highly accurate tool to measure the cluster mass. The massive
gravity of the causes the light from the background objects to bend around the cluster
and form distorted arc-shaped images. The mass of the cluster can be measured from
the geometry of the distortions. Such measurements have been found to be in

6
agreement with the dynamical mass estimates [9]. The distribution of dark matter
within and around the cluster has also been deduced from the geometry of the
distortions caused by gravitational lensing.

1.4 Aim of the Study

The aim of the project is to determine the velocity dispersion and, subsequently, the
dynamical mass of galaxy clusters taken from the GMBCG cluster catalog using virial
theorem. The corresponding data for each of the galaxy clusters that is studied are
obtained from the Sloan Digital Sky Survey (SDSS) database. Following this, the
dynamical mass is compared with a rough estimate of the luminous mass, which
reveals that the dynamical mass is larger than the luminous mass by a few orders.
Further, the crossing time is calculated from the average velocity of the member
galaxies obtained from the calculations, and then compared with the known age of the
universe, which shows that the galaxies within the cluster are held together in a
virialized system over a large time scale. These discrepancies can only be resolved by
assuming that the source of the additional gravitational support that holds the galaxies
in the cluster together in space is a source of mass, whose properties are unknown,
called the dark matter.

1.5 Scope of the Study

The study of clusters is key to understanding dark matter and dark energy in the
universe. Galaxy clusters and dark matter are two very young areas of research in the
fields of astrophysics and cosmology. Cluster masses are being discovered with more
accuracy using gravitational lensing effect and X-ray imaging. Although dynamical
mass estimates only hold in so far as the cluster under consideration is stable and is in
a virialized state, several methods have shown to corroborate these masses for low to
medium redshift clusters. X-ray and lensing studies of clusters of galaxies have the
potential to provide more information about their dynamical masses, which can be
compared to the virial estimates. On more than one occasion, the dynamical masses
estimated from X-ray and lensing studies have been found to be in agreement with the
dynamical mass estimated from the velocity dispersion by the application of virial
theorem.

7
1.6 Limitations of the Study

The main limitation of the virial theorem method is that, though we are able to
establish the presence of dark matter in the clusters, there is no way to explain or
obtain any knowledge about the distribution of the dark matter within the clusters by
using this method alone. Application of the virial theorem implicitly assumes that
only one potential is being probed by the characteristic velocity dispersion. If clusters
have parts which have not been assimilated into the main potential, then there are
multiple potentials and the observed value of characteristic velocity dispersion is not
valid in the context of the virial theorem. Secondly, the method relies heavily on the
assumption that the clusters are stable. While this assumption holds well for low
redshift clusters, the same cannot be said about other clusters. It also depends on the
assumption that the mass follows the light. However, recent observations show that
the virial masses are in good agreement with the masses obtained from lensing
methods [10]. If this is true, then the masses determined by the virial method are
accurate [11].

8
 Chapter 2

REVIEW OF LITERATURE

Zwicky calculated the radial velocity dispersions of 800 galaxies in the Coma Cluster
from their Doppler redshifts and applied the virial theorem to it by assuming the
cluster had reached a mechanically stable state to estimate its dynamical mass. He
assumed that the mass of a galaxy within the clusters as determined from its
luminosity was approximately 109 M☉ so that the mass of the cluster would be 8
1011 M☉. But the velocity dispersion derived from this value was 80 km/s, which was
much lower than the observed dispersions. The observed spread of the velocities
ranged between 1500 and 2000 km/s and the average radial velocity of the cluster was
determined to be close to 1000 km/s. He concluded that in order for this to be true,
the average density of matter present in the Coma system would have to be at least
400 times greater than that derived from the observation of luminous matter, and
hypothesized that, without a substantial source of gravity, the observed velocities of
the cluster (averaging about 1000 km/s) would cause the Coma cluster to fly apart. He
concluded that dunkle Materie or dark matter would have to be present in much
greater density than luminous matter. [12]

Sinclair Smith studied the radial or line-of-sight velocities of thirty two galaxies in the
Virgo Cluster. By examining the distribution of velocities of the cluster in space, he
was able to establish that there was no accumulation of high velocities towards the
centre of the cluster which meant there was no grouping or evaporation occurring.
Hence, he assumed that the cluster was in a relatively stable state held together by its
own gravity. The mass of the cluster as determined from the velocity distribution was
found to be around 1014 M☉, which was much larger than the value estimated by
Hubble of around 109 M☉. He stated that it was possible that both these values were
correct and that the apparent difference could be owed to the ―great mass of
internebular material within the cluster.‖ The existence of the strong gravitational
field of the cluster was also supported by the individual velocities of the nebulae or
galaxies in the cluster which were four times as high as the values determined
previously. [13]
9
In his paper titled ‗On the Masses of Nebulae and of Clusters of Nebulae,‘ Zwicky
discusses new methods of determining the masses of nebulae and clusters. Again, he
assumed that the Coma Cluster whose radius was measured to be 2 106 light years
was a mechanically stable system and applied the virial theorem to it and found that
the average mass of the individual galaxies of the Coma Clusters were at least 4.5
1011 M☉. [14]

The masses of four intermediate redshift galaxy clusters are derived from
gravitational lensing effect on the total mass of the cluster. This method takes the help
of the arc-like images of the background objects and/or galaxies, which allows one to
calculate the gravitational effects of the foreground cluster independent of the state of
the cluster and regardless of whether or not it is in hydrostatic equilibrium. They were
compared to the dynamical masses determined from applying virial theorem to the
systems. It was found that the dynamical masses fell approximately 3 – 6 times short
of the lensing mass within a radius of approximately 0.3Mpc h-1. The decreased
dynamical mass leads to the conclusion that the clusters may not always be assumed
to be well-relaxed virialized systems. [15]

Ragnvald J. Irgens and Per B. Lilje calculated the dynamical velocity dispersions and
dynamical masses for a sample of X-ray luminous rich galaxy clusters at intermediate
redshifts (around z = 0.2 to 0.3) which were drawn from sample of 39 clusters. The
gravitational lens mass of these clusters were also estimated. The velocity dispersions
were measured from 9 to 20 redshifts with the low-dispersion survey spectrograph of
the the William Herschel Telescope, as well as their virial radii were determined from
the imaging with the help og the UH8K mosaic CCD camera on the University of
Hawaii 2.24 m telescope. The velocity dispersions for 12 clusters in the gravitational
lensing sample were determined and compared the dispersion estimates with our
estimates of the velocity dispersions made from gravitational lensing. In all but two
clusters, good agreement between the velocity dispersion estimates based on and
those based on weak lensing. The good agreement that we find between virial masses
and lensing masses for most of our clusters is consistent with the picture that that the
majority of massive clusters at z = 0.2 – 0.3 are close to dynamic equilibrium. [16]

10
 Chapter 3

DETERMINATION OF DYNAMICAL MASS OF

GALAXY CLUSTERS

3.1 Introduction

The determination of galaxy cluster mass is of great importance since it is directly

linked to the well- known problem of dark matter in the Universe and to the cluster
baryon content [11]. In 1933, Zwicky showed that the virialized or dynamical mass of
the Coma Cluster exceeds the luminous mass by a large factor. Similar results were
also obtained for the Virgo Cluster by Smith in 1936. In this chapter, the method for
determining the dynamical mass of galaxy clusters by the application of virial
theorem is explained in detail. It involves using a Python programme to calculate the
velocity dispersions of the individual galaxies in the cluster, as well as estimating the
radius of the cluster. These are then plugged into the virial theorem equation to obtain
the mass of the cluster.

3.2 The Virial Theorem Method

Structures in the universe form hierarchically i.e., small sub galactic bodies first come
together under gravity and collapse to form a stable system, such as a galaxy. Several
galaxies that exist together in the same region will further collapse over time under
the gravitational fields of individual galaxies to form a group or cluster of galaxies.
Since these processes occur over a large time scale, the clusters can be assumed to be
virialized systems, and the virial theorem can be applied to them to calculate their
dynamical masses.

The virial theorem in mechanics relates the average kinetic energy of a system of
bound particles in stable equilibrium with the average potential energy of the system.
The theorem is applicable to complex gravitating systems such as galaxies and
clusters of galaxies, for which an exact solution is difficult to calculate.

11
Mathematically, it is expressed as
………… (3.1)
………… (3.2)
where is the average kinetic energy; and
is the potential energy of the system of N particles.

In the case of this project, the particles are replaced by the member galaxies that are
held stably within the cluster by the gravitational potential of the cluster.

is the average kinetic energy in terms of the characteristic velocity dispersion of the
galaxies in the cluster, and it is expressed as

………… (3.3)

where is the mass of individual galaxies.

is the gravitational potential energy of the cluster that binds the galaxies together
into a virialized system within the cluster. It is expressed as

………… (3.4)

where 6.673×10-11 N m2 kg-2 is the gravitational constant; and

is the radius of the cluster

From the above equations, the expression for the dynamical mass of the cluster can be
determined, and is given by

………… (3.5)

3.2.1 Characteristic Velocity Dispersion

The average kinetic energy of the cluster of galaxies is calculated from the
characteristic value of velocity dispersion of the individual galaxies in the cluster. The
velocity dispersion is the dispersion of the radial velocities of the individual galaxies
about the mean velocity of the cluster.

12
The radial velocity is determined from the spectroscopic redshifts of the individual
galaxies. Redshift refers to the shift in the wavelength of radiation coming from an
object that is moving away from the observer towards the red end of the
electromagnetic spectrum (i.e., increase in wavelength) due to the Doppler Effect. It is
given by the expression,

………… (3.6)

If we take as the radial velocity of each galaxy in the cluster, then the above
equation can be expressed as follows,

⁄
√
⁄
………… (3.7)

or ………… (3.8)

If is the mean redshift of the cluster and as the mean radial velocity of the cluster,
then the velocity dispersion of individual galaxies will be given by,

………… (3.9)

i.e., the velocity dispersion of the ith galaxy is given by,

………… (3.10)

Then the characteristic or root mean square (rms) value of the velocity dispersion of
the cluster will be,

√∑ ………… (3.11)

where is the total number of galaxies in the cluster.

3.2.2 Estimating the Size of the Clusters

The size or width of the cluster is estimated by considering the entire range of angular
separations between the galaxies in the cluster and selecting the largest separation
between two galaxies within the cluster after carefully eliminating outliers. This is
achieved by using the law of cosines in spherical coordinates (explained in section

13
3.2.3) to compute the angular separation between each galaxy with every other galaxy
in the cluster. Then, a histogram is plotted showing the distribution of the separations,
taking angular separation on the x-axis and frequency of angular separation on the y-
axis. From the graph, a suitable value of angular separation is selected as the
approximate width of the cluster. The apparent angular separation is then converted
into a physical distance by using the angular diameter distance (explained in section
3.2.4) to the cluster.

3.2.3 Law of Cosines in Spherical Coordinates

The law of cosines in spherical coordinates relates the sides and the angles of a
triangle on a spherical surface by the equation,

………… (3.12)

Figure 3.1 Spherical triangle solved by the law of cosines.

In Figure 3.1, is the distance between the points v and w, and it can be obtained by
taking the arc cosine of the value of the right hand side of the equation.

In astronomy, we use a two dimensional spherical coordinate system called the

equatorial system, which maps the sky using two coordinates, Right Ascension (RA)
and Declination (Dec). The Equatorial system projects the equator and poles of the
Earth onto the celestial sphere as shown in Figure 3.2. The Right Ascension is
measured east along the celestial equator and Declination is measured north or south
from the celestial equator. The position of an object in the sky will be represented in
the equatorial system as the ordered pair (RA, Dec) or .

14
 Figure 3.2 Definition of equatorial coordinates system on the sky. Observer is at the center of the
sphere, which has an arbitrary radius of ∞ [17].

To calculate the distance between two points (or objects) in the sky, we use a
modified version of the spherical law of cosines, given by

………… (3.13)

where is the RA coordinate of the first object;

is the Dec coordinate of the first object;

is the RA coordinate of the second object;

is the Dec coordinate of the second object; and

is the angular distance of separation between the two objects.

3.2.4 Angular Diameter Distance ( )

Since it is not possible to make direct measurements of the physical sizes of celestial
objects in astronomy and cosmology, we have to rely on other quantities that can be
measured directly from Earth. It is possible to measure an object‘s angular size, with
which we can to estimate their actual size using a quantity called the angular diameter
distance. Angular diameter distance ( to an object is a measure of distance, which
15
is defined in terms of the angular size of the object as viewed from the Earth and its
actual physical size.

In general it is expressed as

………… (3.14)

where is the physical size of the object; and

is the angular size of the object.

However, in the current accepted cosmology, which favours the ΛCDM model, the
relation is slightly more complicated and depends on the Hubble‘s constant, two
cosmological parameters, and the redshift of the cluster. It is expressed as

∫ ………… (3.15)
√

where = 3 105 km/s is the speed of light in air or vacuum;

km/s/Mpc is the Hubble‘s constant;

0.27 is the matter density parameter;

0.73 is the dark energy density parameter; and

is the mean redshift of the cluster.

This expression takes into account the expansion rate and flatness of the universe.

16
 Chapter 4

DATA ANALYSIS

4.1 Introduction

The galaxy clusters studied in this project were selected from the GMBCG Cluster
catalog for SDSS DR 7 and the corresponding data were obtained from the Sloan
Digital Sky Survey (SDSS) Science Archive Server. The dynamical mass of the
cluster was calculated using a python programme designed to compute by calculating
the velocity dispersion of individual galaxies and approximate size of the cluster from
the SDSS data and applying to the virial mass equation. The same programme also
calculates the crossing time scale for the clusters. The data analyses and results for the
dynamical mass calculations of three galaxy clusters have been elaborated in this
chapter, along with some discussion about SDSS, the GMBCG Catalogue and the
Python modules used.

4.2 Sloan Digital Sky Survey (SDSS)

The Sloan Digital Sky Survey (SDSS) is one of the most ambitious and influential
surveys in the history of astronomy. Over eight years of operations (SDSS-I, 2000-
2005; SDSS-II, 2005-2008), the elaborate survey mapped one third of the sky by
performing redshift surveys of galaxies, quasars and stars. It has produced deep,
digitized multi-colour CCD images, and corresponding spectra for more than three
million astronomical objects including 930,000 galaxies and more than 120,000
quasars, making it the most detailed three-dimensional maps of the universe ever
made. [29]

The SDSS consists of a series of three interlocking imaging and spectroscopic

surveys, carried out over an eight-year period. It uses a dedicated 2.5m telescope
located at the Apache Point Observatory (APO) in Southern New Mexico, United
States. The 120-megapixel camera imaged 1.5 square degrees of sky at a time, about
eight times the area of the full moon. A pair of spectrographs fed by optical fibers
measured spectra of (and hence distances to) more than 600 galaxies and quasars in a
17
single observation. A custom-designed set of software pipelines kept pace with the
enormous data flow from the telescope. The two key technologies that enabled the
SDSS, optical fibers and the digital imaging detectors known as CCDs, were
discoveries that were awarded the 2009 Nobel Prize in Physics.

During its first phase of operations, 2000-2005, the SDSS imaged more than 8,000
square degrees of the sky in five optical bandpasses, and it obtained spectra of
galaxies and quasars selected from 5,700 square degrees of that imaging. It also
obtained repeated imaging (roughly 30 scans) of a 300 square degree stripe in the
southern Galactic cap.

With new financial support and an expanded collaboration including 25 institutions

around the globe, SDSS-II carried out three distinct surveys:

 The Sloan Legacy Survey completed the original SDSS imaging and
spectroscopic goals. The final dataset includes 230 million celestial objects
detected in 8,400 square degrees of imaging and spectra of 930,000 galaxies,
120,000 quasars, and 225,000 stars.
 SEGUE (the Sloan Extension for Galactic Understanding and Exploration)
probed the structure and history of the Milky Way galaxy, with new imaging
of 3500 square degrees and spectra of 240,000 stars in a variety of categories
in selected fields.
 The Sloan Supernova Survey carried out repeat imaging of the 300 square
degree southern equatorial stripe to discover and measure supernovae and
other variable objects. In the course of three 3-month campaigns, the
supernova survey discovered nearly 500 spectroscopically confirmed Type Ia
supernovae, which are being used to determine the history of the accelerating
cosmic expansion over the last 4 billion years.

SDSS data have been released to the scientific community and the general public in
annual increments, and the final data release from SDSS-II being Data release 7
which was released in October 2008. The third and fourth phases of the survey,
SDSS-III and SDSS-IV, have continued to release new data in subsequent with the
latest being DR 16 from SDSS-IV. The final release from SDSS-IV is scheduled for
July 2021, and the fifth phase SDSS-V is scheduled to release the first set of data in
two years. [29]
18
4.2.1 SDSS Data Release 7

Data Release 7 (DR7) is the seventh major data release of the Sloan Digital Sky
Survey and it made available images, imaging catalogs, spectra, and redshifts to the
public and scientific communities for download. It is the final data release of SDSS-
II, an extension of the original SDSS project consisting of three subprojects: The
Legacy Survey, SEGUE and a Supernova survey and it marked the completion of the
SDSS‘s original goals.

The release includes 11663 square degrees of imaging data and contains five-band
photometry for 357 million distinct objects. There are over 1.6 million spectra of
930,000 galaxies, 120,000 quasars, and 460,000 stars [18], in total.

4.2.2 GMBCG Cluster Catalog

In astronomy, a catalog is a collection of data about astronomical objects which are

grouped together based on common properties and morphologies. The early cluster
catalogs that are available such as the Abell Cluster Catalog [20] were constructed
completely by visually inspecting photographic plates. This is an extremely time-
consuming process; moreover, the human eye, despite its efficiency and accuracy,
suffers from subjectivity and incompleteness [21]. Nowadays, catalogs are created on
a large scale by complex computer algorithms that scan the data to accurately separate
out different kinds of objects using specific selection functions.

The Gaussian Mixture Brightest Cluster Galaxy (GMBCG) cluster catalog for SDSS
DR7 covers the Legacy Survey area of SDSS and spans a redshift ranging from 0.1 to
0.55. Its goal was to detect galaxies clustered in three spatial dimensions, using the
information which is available only in two dimensions, i.e., the RA and Dec values.
The catalog consists of over 55,400 rich clusters which were optically selected by
applying the GMBCG algorithm to 8240 square degrees of photometric data [22]
from SDSS Data Release 7. It is one of the largest optical galaxy cluster catalogs.

The GMBCG cluster catalog is approximately volume limited up to redshift z ~ 0.4

and showed a high level of purity and completeness when tested against a mock
catalog and is an extension of the maxBCG cluster catalog for SDSS to redshift
beyond 0.3 and covers slightly more sky area that it.

19
The GMBCG algorithm detects clusters by looking at the difference between cluster
and background in color space, rather than matching the assumed model cluster filter
to select clusters as in the maxBCG algorithm. more than 7,500 square degrees of the
North Galactic Cap, and three stripes in the South Galactic Cap totaling 740 square
degrees [18].

4.3 Python

Python is a high-level, interpreted, object-oriented, general purpose programming

language that is widely used by scientific applications and other fields such as
mathematics and engineering. It was created by Guido van Rossum and released in
1991. Python's design philosophy emphasizes code readability with its notable use of
significant whitespace. Its language constructs and object-oriented approach aim to
help programmers write clear, logical code for small and large-scale projects [23].
The latest version of python is Python 3.8.3; however, for this project, the data
analysis was done using Python 3.7.4 with the help of some of the libraries available
in the SciPy package.

4.3.1 SciPy

SciPy is a Python-based ecosystem of open-source software for mathematics, science,

and engineering [24]. It has various libraries available, which are highly useful for
scientific computing. These include NumPy, Maplotlib, and pandas libraries, which
have been used in this project, along with SymPy and IPython.

The NumPy, or Numerical Python, library was created specifically for numerical
computing in the form of N-dimensional base arrays. The Matplotlib library can be
used for effective 2-dimensional graph plotting as well as creating static, animated,
and interactive visualizations in Python [25]. It was used to plot histograms to
estimate the size of the clusters in this project. The pandas library is a useful tool for
fast and efficient for data manipulation with integrated indexing, and for reading and
writing data in different formats such as CSV files, as in this particular project.

20
4.3.2 CSV Module
The data for each cluster obtained from the SDSS database was exported as Comma
Separated Value (CSV) files with .csv extension. The csv module in Python
implements classes and objects to read and write tabular data in CSV format [26]. The
main objects used in this project for analysing the CSV data files were the reader and
writer objects which were used to read data from RA, Dec and Redshift columns
from the CSV file and write them to separate CSV files for processing using the
Python [27] programme to calculate dynamical mass.

4.4 Data Analysis

The galaxy clusters studied in this project were selected from the GMBCG Cluster
catalog for SDSS DR 7 and the corresponding data were obtained from the Sloan
Digital Sky Survey (SDSS) Science Archive Server. To retrieve data about the
member galaxies of the cluster, the RA and Dec coordinates and redshift range of the
cluster and a search radius of 0.3 degrees were entered in the search tool of the SDSS
SAS database. The search yields a table containing the RA and Dec values as well as
the spectroscopic redshifts of the member galaxies of the cluster. The tabulated data
can be exported as a CSV (comma separated value) file from the database.

The python programmes, given in Appendices A.1, A.2 and A.3 for the three galaxy
clusters analyzed in this chapter respectively, computes the dynamical mass of the
cluster by calculating the velocity dispersion of individual galaxies and approximate
characteristic of the cluster from the data and applying to the virial mass equation
given in equation 3.11. The characteristic radius is selected from a histogram showing
angular separation distribution over the entire range of galaxies in the cluster. The
programme also calculates the crossing time scale for the clusters.

21
 Figure 4.1 An SDSS Image of Galaxy Cluster GMBCG J197.87292-01.34109 with (RA, Dec) =
(197.87292, -01.34109) and redshift z = 0.203.

4.4.1 Data Analysis of Cluster GMBCG J197.87292-01.34109

The cluster‘s (RA, Dec) coordinates are (197.87292, -01.34109) and its redshift is
0.203. Figure 4.1 shows an image of the cluster taken from the SDSS database. A
minimum redshift of 0.18 and maximum redshift of 0.23, and search radius 0.3 were
entered into the Optical Spectra Advanced Search tool of the SDSS DR12 Science
Archive Server (SAS) database [28]. The table of data of individual galaxies in the
cluster was exported as a CSV file. The python programme was executed and it
calculated the dynamical mass of the galaxy cluster from the characteristic velocity
dispersion and radius of the cluster. The radius of the cluster was selected as 0.35
degrees from the histogram shown in Figure 4.2 after carefully eliminating outliers.
22
The programme also calculated the crossing time scale for the cluster. The results for
the cluster are detailed below.

Figure 4.2 Histogram showing distribution of angular separations for all possible pairs of galaxies in
the Cluster GMBCG J197.87292-01.34109. The size of the cluster was chosen as 0.35 after eliminating
extreme values.

4.4.2 Results for Cluster GMBCG J197.87292-01.34109

The mean redshift of the cluster was found to be approximately 0.1892. The velocity
dispersion of the cluster was approximately 3492.78 km/s and the angular diameter
distance was approximately 2.03 1022 km or 658.5 Mpc. The angular size of the
cluster was selected as 0. 35 degrees from the histogram shown in Figure 4.2. The
physical radius of the cluster was calculated using the angular separation and the
angular diameter distance and was found to be approximately 6.206 1019 km or
2.01 Mpc. Thy dynamical mass of the cluster was found to be approximately 3.78
1045 kg or 1.9 1015 . Finally, the crossing time scale for the cluster was found to
be 3.46 107 years.

23
 Figure 4.3 An SDSS Image of Galaxy Cluster GMBCG J231.03093+29.88897 with (RA, Dec) =
(231.03093, 29.88897) and redshift z = 0.114.

4.4.3 Data Analysis of Cluster GMBCG J231.03093+29.88897

The cluster‘s (RA, Dec) coordinates are (231.03093, 29.88897) and its redshift is
0.114. Figure 4.3 shows an image of the cluster taken from the SDSS database. A
minimum redshift of 0.084 and maximum redshift of 0.134, and search radius 0.3
were entered into the Optical Spectra Advanced Search tool of the SDSS DR12
Science Archive Server (SAS) database [28]. The table of data of individual galaxies
in the cluster was exported as a CSV file. The python programme was executed and it
calculated the dynamical mass of the galaxy cluster from the characteristic velocity
dispersion and radius of the cluster. The radius of the cluster was selected as 0.4
degrees from the histogram shown in Figure 4.4 after carefully eliminating outliers.

24
The programme also calculated the crossing time scale for the cluster. The results for
each cluster are detailed below.

Figure 4.4 Histogram showing distribution of angular separations for all possible pairs of galaxies in
the Cluster GMBCG J231.03093+29.88897. The size of the cluster was chosen as 0.4 after eliminating
extreme values.

4.4.4 Results for Cluster GMBCG J231.03093+29.88897

The mean redshift of the cluster was found to be approximately 0.1143. The velocity
dispersion of the cluster was approximately 1939.44 km/s and the angular diameter
distance was approximately 1.33 1022 km or 431.7 Mpc. The angular size of the
cluster was selected as 0.4 degrees from the histogram shown in Figure 4.5. The
physical size of the cluster was calculated using the angular separation and the
angular diameter distance and was found to be approximately 4.65 1019 km or 1.51
Mpc. Thy dynamical mass of the cluster was found to be approximately 8.73 1044
kg or 4.39 1014 M☉. Finally, the crossing time scale for the cluster was found to be
4.67 107 years.

25
 Figure 4.5 An SDSS Image of Galaxy Cluster GMBCG J47.12599 -0.353214 with (RA, Dec) =
(47.12599, -0.353214) and redshift z = 0.374.

4.4.5 Data Analysis of Cluster GMBCG J47.12599 -0.353214

The cluster‘s (RA, Dec) coordinates are (47.12599, -0.353214) and its redshift is
0.374. Figure 4.5 shows an image of the cluster taken from the SDSS database. A
minimum redshift of 0.344 and maximum redshift of 0.404, and search radius 0.3
were entered into the Optical Spectra Advanced Search tool of the SDSS DR12
Science Archive Server (SAS) database [28]. The table of data of individual galaxies
in the cluster was exported as a CSV file. The python programme was executed and it
calculated the dynamical mass of the galaxy cluster from the characteristic velocity
dispersion and radius of the cluster. The radius of the cluster was selected as 0.45
degrees from the histogram shown in Figure 4.6 after carefully eliminating outliers.
26
The programme also calculated the crossing time scale for the cluster. The results for
each cluster are detailed below.

Figure 4.6 Histogram showing distribution of angular separations for all possible pairs of galaxies in
the Cluster GMBCG J47.12599 -0.353214. The size of the cluster was chosen as 0.45 after eliminating
extreme values.

4.4.6 Results for Cluster GMBCG J47.12599 -0.353214

The mean redshift of the cluster was found to be approximately 0.3771. The velocity
dispersion of the cluster was approximately 4647 km/s and the angular diameter
distance was approximately 3.346 1022 km or 1084.42 Mpc. The angular size of the
cluster was selected as 0.45 degrees from the histogram shown in Figure 4.6. The
physical size of the cluster was calculated using the angular separation and the
angular diameter distance and was found to be approximately 1.314 1020 km or
4.26 Mpc. Thy dynamical mass of the cluster was found to be approximately 1.417
1046 kg or 7.125 1015 M☉. Finally, the crossing time scale for the cluster was found
to be 4.55 107 years.

27
4.5 Conclusions

It is clear that the dynamical mass is significantly higher than the average luminous
mass of galaxy clusters. For the clusters GMBCG J197.87292-01.34109 and GMBCG
J231.03093+29.88897, the dynamical masses were of the order ~1015 M☉, which is
about three orders larger than the typical luminous mass of clusters. The dynamical
mass of the cluster J47.12599 -0.353214 was of the order of ~1014 M☉, which is two
orders higher than the typical cluster mass (~1012 M☉). Thus, it can be inferred that
the cluster dynamics is dominated by dark matter. The crossing time scale was also
found to be significantly smaller than the current estimate of the age of the universe
which is about 13.7 billion years or 1.37 1010 years. The crossing time scale
estimated for all three clusters were of the order of ~ 107 years, which is almost three
orders lesser than the age of the universe. Thus, it is safe to assume that the galaxies
within the cluster have had enough time to enter a stable, virialized state in the cluster
and the application of virial theorem to calculate the mass of this cluster is valid.

28
 Appendix A
PYTHON CODES

A.1 Python Code for Cluster GMBCG J197.87292-01.34109

import csv
import math
import numpy as np
import pandas as pd
import [Link] as plt
from [Link] import quad

# extract and write only redshift column to new csv file designated
as 'filename'

source = '[Link]'
filename = '[Link]'

def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(filename, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 7
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(filename, 'r') as csv_file:

[Link](csv_file)

extract(filename) # writes redshift column of [Link] file into

new csv file, [Link]

def readcsv(filename): # convert column to array

data = pd.read_csv(filename)
return [Link](data)

z = readcsv(filename) # assigns redshift array to a variable z

# print("The redshifts of the galaxies are:", "\n", z)
zcluster = [Link](z) # take mean of redshifts (array)
print("\n The mean redshift of the cluster is", zcluster)

29
c = 3 * (10 ** 5) # speed of light in km/s
u = (((1 + zcluster) ** 2 - 1) / ((1 + zcluster) ** 2 + 1)) * c #
mean velocity of the cluster
n = len(z) # number of galaxies in the cluster

sum = 0

for i in range(n):
v = (((1 + z[i]) ** 2 - 1) / ((1 + z[i]) ** 2 + 1)) * c
disp = (u - v) # velocity dispersion
# print(disp)
y = disp ** 2
sum = sum + y
i += 1

vrms = [Link](sum / n)
# print("The rms value of velocity dispersion is", vrms)
print("\n The characteristic velocity dispersion of the cluster is",
[Link](3) * vrms, "km/s")

ra = '[Link]'

def extract(ra):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(ra, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 5
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(ra, 'r') as csv_file:

z = [Link](csv_file)

extract(ra)

def readcsv(ra): # convert column to array

data = pd.read_csv(ra)
return [Link](data)

rtasc = readcsv(ra) # RA values

dec = '[Link]'

30
def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(dec, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 6
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(dec, 'r') as csv_file:

z = [Link](csv_file)

extract(dec)

def readcsv(dec): # convert column to array

data = pd.read_csv(dec)
return [Link](data)

decl = readcsv(dec) # Dec values

theta = [] # angular separations

for i in range(len(decl)):

one = [Link]([Link](90 - decl[i]))

two = [Link]([Link](90 - decl[i]))

for j in range(len(decl)):
if i < j:

three = [Link]([Link](90 - decl[j]))

four = [Link]([Link](90 - decl[j]))

for k in range(len(rtasc)):
for l in range(len(rtasc)):

if k < l:
five = [Link]([Link](rtasc[k] -
rtasc[l]))
six = one * three + two * four * five

if six > 1:
seven = [Link]([Link](1))
else:
seven = [Link]([Link](six))

31
 if i == k:
if j == l:
# print(i, j, k, l, six, seven)
[Link](seven)

[Link](figsize=(16, 10))
[Link]('xtick', labelsize=14)
[Link]('ytick', labelsize=14)
[Link]('font', size=16)
[Link]('Number of Galaxy Pairs')
[Link]('Angular Separation')
[Link](theta, bins=20)

def f(w):
return 1 / ([Link]((0.27 * (1 + w) ** 3) + 0.73))

intg = quad(f, 0, zcluster)

h0 = 69.6 # Hubble's constant
integral = intg[0]
# print (integral)
const = c / (h0 * (1 + zcluster))
da = (c / (h0 * (1 + zcluster))) * integral
dakm = da * (3.08567758128 * (10 ** 19))
print('\nThe angular diameter distance is', da, 'Mpc or', dakm,
'km') # Angular diameter distance

radius = dakm * ([Link]([Link](0.35 / 2)))

print('\nThe radius of the cluster is', radius, 'km')

gconst = 6.67408 * (10 ** (-11))

r = radius * (10 ** 3) # Radius in metre
vrms1 = vrms * (10 ** 3)
# print (vrms, vrms1)
dynmass = (r * (vrms1 ** 2)) / gconst
print('\nThe dynamical mass of the cluster is', dynmass, 'kg')

t = (radius / v)
tcross = t / 31536000
print('The crossing time of the cluster is ', tcross)

32
A.2 Python Code for Cluster GMBCG J231.03093+29.88897
import csv
import math
import numpy as np
import pandas as pd
import [Link] as plt
from [Link] import quad

# extract and write only redshift column to new csv file designated
as 'filename'

source = '[Link]'
filename = '[Link]'

def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(filename, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 7
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(filename, 'r') as csv_file:

[Link](csv_file)

extract(filename) # writes redshift column of [Link] file into

new csv file, [Link]

def readcsv(filename): # convert column to array

data = pd.read_csv(filename)
return [Link](data)

z = readcsv(filename) # assigns redshift array to a variable z

# print("The redshifts of the galaxies are:", "\n", z)
zcluster = [Link](z) # take mean of redshifts (array)
print("\n The mean redshift of the cluster is", zcluster)

c = 3 * (10 ** 5) # speed of light in km/s

u = (((1 + zcluster) ** 2 - 1) / ((1 + zcluster) ** 2 + 1)) * c #
mean velocity of the cluster
n = len(z) # number of galaxies in the cluster

sum = 0

33
for i in range(n):
v = (((1 + z[i]) ** 2 - 1) / ((1 + z[i]) ** 2 + 1)) * c
disp = (u - v) # velocity dispersion
# print(disp)
y = disp ** 2
sum = sum + y
i += 1

vrms = [Link](sum / n)
# print("The rms value of velocity dispersion is", vrms)
print("\n The characteristic velocity dispersion of the cluster is",
[Link](3) * vrms, "km/s")

ra = '[Link]'

def extract(ra):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(ra, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 5
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(ra, 'r') as csv_file:

z = [Link](csv_file)

extract(ra)

def readcsv(ra): # convert column to array

data = pd.read_csv(ra)
return [Link](data)

rtasc = readcsv(ra) # RA values

dec = '[Link]'

def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(dec, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

34
 t = 6
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(dec, 'r') as csv_file:

z = [Link](csv_file)

extract(dec)

def readcsv(dec): # convert column to array

data = pd.read_csv(dec)
return [Link](data)

decl = readcsv(dec) # Dec values

theta = [] # angular separations

for i in range(len(decl)):

one = [Link]([Link](90 - decl[i]))

two = [Link]([Link](90 - decl[i]))

for j in range(len(decl)):
if i < j:

three = [Link]([Link](90 - decl[j]))

four = [Link]([Link](90 - decl[j]))

for k in range(len(rtasc)):
for l in range(len(rtasc)):

if k < l:
five = [Link]([Link](rtasc[k] -
rtasc[l]))
six = one * three + two * four * five

if six > 1:
seven = [Link]([Link](1))
else:
seven = [Link]([Link](six))

if i == k:
if j == l:
# print(i, j, k, l, six, seven)
[Link](seven)

[Link](figsize=(16, 10))
[Link]('xtick', labelsize=14)
[Link]('ytick', labelsize=14)

35
[Link]('font', size=16)
[Link]('Number of Galaxy Pairs')
[Link]('Angular Separation')
[Link](theta, bins=20)

def f(w):
return 1 / ([Link]((0.27 * (1 + w) ** 3) + 0.73))

intg = quad(f, 0, zcluster)

h0 = 69.6 # Hubble's constant
integral = intg[0]
# print (integral)
const = c / (h0 * (1 + zcluster))
da = (c / (h0 * (1 + zcluster))) * integral
dakm = da * (3.08567758128 * (10 ** 19))
print('\nThe angular diameter distance is', da, 'Mpc or', dakm,
'km') # Angular diameter distance

angsize = float(input('Select a suitable angular size from

histogram: '))
radius = dakm * ([Link]([Link](0.4 / 2)))
print('\nThe radius of the cluster is', radius, 'km')

gconst = 6.67408 * (10 ** (-11))

r = radius * (10 ** 3) # Radius in metre
vrms1 = vrms * (10 ** 3)
# print (vrms, vrms1)
dynmass = (r * (vrms1 ** 2)) / gconst
print('\nThe dynamical mass of the cluster is', dynmass, 'kg')

t = (radius / v)
tcross = t / 31536000
print('The crossing time of the cluster is ', tcross)

36
A.3 Python Code for Cluster GMBCG J47.12599 -0.353214
import csv
import math
import numpy as np
import pandas as pd
import [Link] as plt
from [Link] import quad

# extract and write only redshift column to new csv file designated
as 'filename'

source = '[Link]'
filename = '[Link]'

def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(filename, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 7
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(filename, 'r') as csv_file:

[Link](csv_file)

extract(filename) # writes redshift column of [Link] file into

new csv file, [Link]

def readcsv(filename): # convert column to array

data = pd.read_csv(filename)
return [Link](data)

z = readcsv(filename) # assigns redshift array to a variable z

# print("The redshifts of the galaxies are:", "\n", z)
zcluster = [Link](z) # take mean of redshifts (array)
print("\n The mean redshift of the cluster is", zcluster)

c = 3 * (10 ** 5) # speed of light in km/s

u = (((1 + zcluster) ** 2 - 1) / ((1 + zcluster) ** 2 + 1)) * c #
mean velocity of the cluster
n = len(z) # number of galaxies in the cluster

sum = 0

37
for i in range(n):
v = (((1 + z[i]) ** 2 - 1) / ((1 + z[i]) ** 2 + 1)) * c
disp = (u - v) # velocity dispersion
# print(disp)
y = disp ** 2
sum = sum + y
i += 1

vrms = [Link](sum / n)
# print("The rms value of velocity dispersion is", vrms)
print("\n The characteristic velocity dispersion of the cluster is",
[Link](3) * vrms, "km/s")

ra = '[Link]'

def extract(ra):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(ra, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

t = 5
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(ra, 'r') as csv_file:

z = [Link](csv_file)

extract(ra)

def readcsv(ra): # convert column to array

data = pd.read_csv(ra)
return [Link](data)

rtasc = readcsv(ra) # RA values

dec = '[Link]'

def extract(filename):
with open(source, 'r') as csv_file:
csv_reader = [Link](csv_file)

with open(dec, 'w') as new_file:

csv_writer = [Link](new_file, delimiter=',',
quotechar='"', quoting=csv.QUOTE_MINIMAL)

38
 t = 6
f = t - 1
for line in csv_reader:
csv_writer.writerow(line[f:t])

with open(dec, 'r') as csv_file:

z = [Link](csv_file)

extract(dec)

def readcsv(dec): # convert column to array

data = pd.read_csv(dec)
return [Link](data)

decl = readcsv(dec) # Dec values

theta = [] # angular separations

for i in range(len(decl)):

one = [Link]([Link](90 - decl[i]))

two = [Link]([Link](90 - decl[i]))

for j in range(len(decl)):
if i < j:

three = [Link]([Link](90 - decl[j]))

four = [Link]([Link](90 - decl[j]))

for k in range(len(rtasc)):
for l in range(len(rtasc)):

if k < l:
five = [Link]([Link](rtasc[k] -
rtasc[l]))
six = one * three + two * four * five

if six > 1:
seven = [Link]([Link](1))
else:
seven = [Link]([Link](six))

if i == k:
if j == l:
# print(i, j, k, l, six, seven)
[Link](seven)

[Link](figsize=(16, 10))
[Link]('xtick', labelsize=14)
[Link]('ytick', labelsize=14)

39
[Link]('font', size=16)
[Link]('Number of Galaxy Pairs')
[Link]('Angular Separation')
[Link](theta, bins=20)

def f(w):
return 1 / ([Link]((0.27 * (1 + w) ** 3) + 0.73))

intg = quad(f, 0, zcluster)

h0 = 69.6 # Hubble's constant
integral = intg[0]
# print (integral)
const = c / (h0 * (1 + zcluster))
da = (c / (h0 * (1 + zcluster))) * integral
dakm = da * (3.08567758128 * (10 ** 19))
print('\nThe angular diameter distance is', da, 'Mpc or', dakm,
'km') # Angular diameter distance

angsize = float(input('Select a suitable angular size from

histogram: '))
radius = dakm * ([Link]([Link](0.45 / 2)))
print('\nThe radius of the cluster is', radius, 'km')

gconst = 6.67408 * (10 ** (-11))

r = radius * (10 ** 3) # Radius in metre
vrms1 = vrms * (10 ** 3)
# print (vrms, vrms1)
dynmass = (r * (vrms1 ** 2)) / gconst
print('\nThe dynamical mass of the cluster is', dynmass, 'kg')

t = (radius / v)
tcross = t / 31536000
print('The crossing time of the cluster is ', tcross)

40
 Bibliography

[1] Bahcall, N. A. Clusters and Superclusters of Galaxies. Ann. Rev. Astron.

Astrophys. 26: (1988). p631-86. URL: [Link]

[2] NASA/ JPL-Caltech / L. Jenkins (GSFC). URL:

[Link]
Coma-Cluster

[3] Carlberg, Raymond. ―Galaxy Clusters.‖ Encyclopedia of Astronomy &

Astrophysics: IOP Publishing Ltd. (2006)

[3] Henry, J.P. ―X-Ray Temperatures for the Extended Medium-Sensitivity

Survey High-Redshift Cluster Sample: Constraints on Cosmology and the
Dark Energy Equation of State.‖ ApJ, (2004) p.609, 603–616

[4] Bahcall, N. A. ―Large-Scale Structure In The Universe Indicated By Galaxy

Clusters.‖ New Horizons from Multi-Wavelength Sky Surveys (1996). p. 317.

[5] Myers, S.T., C. Pfrommer, J. Aguirre , J.R. Bond, J.O. Burns, T. Clarke, M.
Devlin, A. Evrard, S. Golwala, S. Habib, K. Heitmann, W.L. Holzapfel, N.E.
Kassim, A. Kravtsov, A.T. Lee, M. Markevich, D. Marrone, D. Nagai, L.
Page, E. Pierpaoli, L. Rudnick, J. Sievers, G. Taylor, M. Voit. ―Galaxy Cluster
Astrophysics and Cosmology: Questions and Opportunities for the Coming
Decade‖ Astro2010 Science White Paper (2009-02-15), p2.

[6] Voit, G. Mark. ―Tracing cosmic evolution with clusters of galaxies.‖ Reviews
Of Modern Physics, 77. (2005) p. 207 – 252.

[7] Lambda-CDM Model. 2019. URL: [Link]

CDM_model

[8] Siegfried, T., ―Hidden space dimensions may permit parallel universes,
explain cosmic mysteries.‖ The Dallas Morning News. (1999).

[9] Wu, X.; Chiueh, T.; Fang, L.; Xue, Y. ―A comparison of different cluster mass
estimates: consistency or discrepancy?‖ Monthly Notices of the Royal

41
 Astronomical Society. 301 (3):(1998). p.861–871. URL: arXiv:astro-
ph/9808179.

[10] Wu, Xiang-Ping & Fang ―A Statistical Comparison of Cluster Mass Estimates
from Optical/X-Ray Observations and Gravitational Lensing.‖ The
Astrophysical Journal, (1997).

[11] Sadat, R. ―Clusters of Galaxies and Mass Estimates.‖ ASP [Link]. 126
(1997) 349.p.3. URL: [Link]

[12] Zwicky, F. ―The redshift of extragalactic nebulae.‖ Helvetica Physica Acta,

Vol 6. (1933). p.110 – 127.

[13] Smith, S. ―The Mass Of The Virgo Cluster.‖ Astrophysical Journal, 83,
(1936): p.23.

[14] Zwicky, F. ―On The Masses Of Nebulae And Of Clusters Of Nebulae.‖ The
Astrophysical Journal: An International Review Of Spectroscopy And
Astronomical Physics (1937): p.217 – 246.

[15] Wu, Xiang-Ping. ―Determining Gravitational Masses of Galaxy Clusters with

Virial Equilibrium and Arc-like Images.‖ The Astrophysical Journal, 436
(1994): p.115 - 118.

[16] Irgens, Ragnvald J. and Per B. Lilje. ―Weak Gravitational Lensing by a

Sample of X-Ray–Luminous Clusters Of Galaxies. II. Comparison with Virial
Masses.‖ The Astrophysical Journal, 579 (2002): p.227–235.

[17] Preparing astronomical observations and observing with OHP facilities -

Scientific Figure on ResearchGate. URL:
[Link]
system-on-the-sky-Observer-is-at-the-center-of-the_fig1_49594075

[18] Abazajian, Kevork , Kevork N. Abazajian, Jennifer K. Adelman-McCarthy,

Marcel A. Ag¨ueros,, Sahar S. Allam,, Carlos Allende, Prieto, Deokkeun An,,
Kurt S. J. Anderson,, Scott F. Anderson, James Annis, Neta A. Bahcall,
C.A.L., Bailer-Jones, J. C. Barentine, Bruce A. Bassett,, Andrew C. Becker,
Timothy C. Beers, Eric F. Bell, Vasily, Belokurov, Andreas A. Berlind, Eileen
F. Berman, Mariangela Bernardi, Steven J. Bickerton, Dmitry, Bizyaev, John
42
P. Blakeslee, Michael R. Blanton, John J. Bochanski,, William N. Boroski,
Howard J. Brewington, Jarle Brinchmann,, J. Brinkmann, Robert J. Brunner,
Tam´as Budav´ari, Larry N. Carey, Samuel Carliles, Michael A. Carr,
Francisco J. Castander, David Cinabro, A. J. Connolly, Istv´an Csabai, Carlos
E. Cunha, Paul C. Czarapata, James R. A. Davenport, Ernst de Haas, Ben
Dilday,,, Mamoru Doi, Daniel J. Eisenstein, Michael L. Evans, N. W. Evans,
Xiaohui Fan, Scott D. Friedman, Joshua A. Frieman, Masataka Fukugita,
Boris T. G¨ansicke, Evalyn Gates, Bruce Gillespie, G. Gilmore, Belinda
Gonzalez, Carlos F. Gonzalez, Eva K. Grebel, James E. Gunn, Zsuzsanna
Gy˝ory, Patrick B. Hall, Paul Harding, Frederick H. Harris, Michael
Harvanek, Suzanne L. Hawley, Jeffrey J.E. Hayes, Timothy M. Heckman,
John S. Hendry, Gregory S. Hennessy, Robert B. Hindsley, J. Hoblitt, Craig J.
Hogan, David W. Hogg, Jon A. Holtzman, Joseph B. Hyde, Shin-ichi
Ichikawa, Takashi Ichikawa, Myungshin Im, ˇZeljko Ivezi´c, Sebastian Jester,
Linhua Jiang, Jennifer A. Johnson, Anders M. Jorgensen, Mario Juri´c,
Stephen M. Kent, R. Kessler, S. J. Kleinman, G. R. Knapp, Kohki Konishi,,
Richard G. Kron,, Jurek Krzesinski,, Nikolay Kuropatkin, Hubert Lampeitl,
Svetlana Lebedeva, Myung Gyoon Lee, Young Sun Lee, R. French Leger,
S´ebastien L´epine, Nolan Li, Marcos Lima, Huan Lin, Daniel C. Long, Craig
P. Loomis, Jon Loveday, Robert H. Lupton, Eugene Magnier, Olena
Malanushenko, Viktor Malanushenko, Rachel Mandelbaum,, Bruce Margon,
John P. Marriner, David Mart´ınez-Delgado, Takahiko Matsubara, Peregrine
M. McGehee, Timothy A. McKay, Avery Meiksin, Heather L. Morrison,
Fergal Mullally, Jeffrey A. Munn, Tara Murphy,, Thomas Nash, Ada Nebot,
Eric H. Neilsen, Jr., Heidi Jo Newberg, Peter R. Newman,, Robert C. Nichol,
Tom Nicinski,, Maria Nieto-Santisteban, Atsuko Nitta, Sadanori Okamura,
Daniel J. Oravetz, Jeremiah P. Ostriker, Russell Owen, Nikhil Padmanabhan,,
Kaike Pan, Changbom Park, George Pauls, John Peoples Jr., Will J. Percival,
Jeffrey R. Pier, Adrian C. Pope, Dimitri Pourbaix,, Paul A. Price, Norbert
Purger, Thomas Quinn, M. Jordan Raddick, Paola Re Fiorentin,, Gordon T.
Richards, Michael W. Richmond, Adam G. Riess, Hans-Walter Rix,
Constance M. Rockosi, Masao Sako,, David J. Schlegel, Donald P. Schneider,
Ralf-Dieter Scholz, Matthias R. Schreiber, Axel D. Schwope, Uroˇs Seljak,,,
Branimir Sesar, Erin Sheldon,, Kazu Shimasaku, Valena C. Sibley, A.E.
43
 Simmons, Thirupathi Sivarani,, J. Allyn Smith, Martin C. Smith, Vernesa
Smolˇci´c, Stephanie A. Snedden, Albert Stebbins Matthias Steinmetz, Chris
Stoughton, Michael A. Strauss, Mark SubbaRao,, Yasushi Suto, Alexander S.
Szalay, Istv´an Szapudi, Paula Szkody, Masayuki Tanaka, Max Tegmark, Luis
F.A. Teodoro, Aniruddha R. Thakar, Christy A. Tremonti, Douglas L. Tucker,
Alan Uomoto, Daniel E. Vanden Berk,, Jan Vandenberg, S. Vidrih, Michael S.
Vogeley, Wolfgang Voges, Nicole P. Vogt, Yogesh Wadadekar,, Shannon
Watters,, David H. Weinberg, Andrew A. West, Simon D.M. White, Brian C.
Wilhite, Alainna C. Wonders, Brian Yanny, D. R. Yocum, Donald G. York,,
Idit Zehavi, Stefano Zibetti, Daniel B. Zucker. ―The Seventh Data Release of
the Sloan Digital Sky Survey.‖ [Link].182 (2009): p. 543-558.
URL: [Link]

[19] Zwicky, F. Catalogue of Galaxies and of Clusters of Galaxies (Pasadena,

California Institute of Technology, 6 vols.). (1961-68).

[20] Abell, G. O., Corwin, H. G. Jr., and Olowin, R. P. ―A Catalog of Rich Clusters
of Galaxies.‖ Astrophys. J. Suppl., 1989, 70, (1989). p1.

[21] Kim, Rita Seung Jung, Jeremy V. Kepner, Marc Postman, Michael A. Strauss,
Neta A. Bahcall, James E. Gunn, Robert H. Lupton, James Annis, Robert C.
Nichol, Francisco J. Castander, J. Brinkmann, Robert J. Brunner, Andrew
Connolly, Istvan Csabai, Robert B. Hindsley, Zˇ eljko Ivezic´, Michael S.
Vogeley, and Donald G. York. ―Detecting Clusters Of Galaxies In The Sloan
Digital Sky Survey. I. Monte Carlo Comparison Of Cluster Detection
Algorithms.‖ The Astronomical Journal, (2002) 123: p.20–36.

[22] Hao, Jiangang, Timothy A. McKay, Benjamin P. Koester, Eli S. Rykoff,

Eduardo Rozo, James Annis, Risa H. Wechsler, August Evrard, Seth R.
Siegel, Matthew Becker, Michael Busha, David Gerdes, David E. Johnston
and Erin Sheldon, ―A GMBCG Galaxy Cluster Catalog of 55,424 Rich
Clusters from SDSS DR7.‖ [Link].191: 254-274. (2010).

[23] Kuhlman, Dave. A Python Book: Beginning Python, Advanced Python, and
Python Exercises. Section 1.1. Archived from the original (PDF) on 23 June
2012.

44
[24] [Link]. URL: [Link]

[25] Matplotlib python plotting. URL: [Link]

[26] CSV File Reading and Writing. URL:

[Link]

[27] Welcome to [Link]. URL: [Link]

[28] Advanced Optical Spectra Search — Optical Spectra — DR12 Science Archive
Server (SAS). URL: [Link]
[29] Sloan Digital Sky Survey. URL: [Link]

45