We now turn our attention to clusters of
galaxies.
Well, they're certainly the most
conspicuous parts of the large scale
structure and were the first to be
recognized as somehow distinct units.
Typically, clusters have sizes of few
megaparsecs and contain hundreds or
thousands of galaxies with masses of 10 to
the 14th to a few times 10 to the 15th
solar masses.
They are all gravitationally bound but
they may not be fully virialized.
As you remember t he free fall time scale
for clusters is measured in gigayears and
so many clusters are still forming.
In addition to the obvious stars and
galaxies, clusters are also filled with a
hot x-ray gas which was partly expelled
from galaxies and partly accreted from the
outside and it's kept on the x-ray
temperatures because of the deep,
potential well.
But the dominant component, of course, is
the dark matter.
Of all galaxies, maybe 10 or 20% are found
in clusters.
The rest are discovering large scale
structures in general, and the majority of
galaxies are really just small groups just
like the local group with Milky Way and
Andromeda dominating a whole bunch of
dwarf galaxies around it.
There is one important distinction though,
the early type galaxies, ellipticals and
S0s tend to concentrate in clusters more
than in general field and there is
relatively few spiral galaxies in
clusters.
This is reflecting processes of the
galaxies evolution and we'll talk about
them a little later.
So here's the picture of the nearest
cluster to us, the Virgo cluster.
It is the center of the local super
cluster.
It's about 15 to 20 megaparsecs away.
The picture on the left is the x-ray
image, in false color.
The picture on the right shows where the
galaxies are.
And you can see there is a broadly
speaking a good respondents.
About 2000 galaxies have been cataloged in
Virgo cluster, but most of them are tiny
dwarfs that you have difficulty actually
pointing out.
The next famous cluster is the coma
cluster, it is five to six times further
away.
And it is richer than Virgo by a
considerable factor, and it's comparable
to the kinds of clusters that we see at
larger distances from us.
Here again, I show you the optical image
in the upper right.
See, there are 2 dominant galaxies there,
and the x-ray image in the lower part and
you can see it's much smoother than the
one for Virgo because Virgo is dynamically
young cluster, still falling together.
Coma is, they're utterly much more
evolved.
The picture on the, on the right just
simply overlays the x-ray countermap on
the optical image.
And on the opposite side of the sky,
roughly, is the Perseus Cluster.
And that is only about 4 or 5 times
further away than Virgo not as rich as
Coma.
But it has a nice little chain of galaxies
in it, one of whom harbors a very powerful
active galactic, nuclea, nucleus.
As we look deeper, we find more clusters
and by and large, some of them look pretty
much like clusters near us.
Again, this is because of cluster
formation time scales are broadly
comprable to the age of the universe, so
we expect to see, more or less, a similar
kind of Clusters at any given time, not
very early because they don't have time to
form it.
This is intermediate distance cluster at
direction 5.43.
Picture on the left is the visible light
image from Hubble space telescope and
picture on the right is the x-ray image
from the Rosat sattelite.
Clusters are now known[UNKNOWN] of maybe 1
and a half or so, and there are hints of
cluster like structures to much higher
that you see it.
.
But this is one of the most distant
clusters proper that we know.
Ah,[unknown] 1.23 and again it looks
reasonably well formed.
So how do we look for clusters?
The most traditional method is just
looking for[UNKNOWN][UNKNOWN] galaxy's on
this The sky.
And this is how most of the initial
catalogs were compiled.
We in fact, we use every one of the
cluster properties as means of finding
them.
So, in the optical, this would be simply
condensed to galaxies.
And nowadays, the way this is done is the
galaxy map on the sky, projected density
map of the sky is smoothed and Algorithm
of some sort is used to identify the other
densities.
Because clusters have a large population
of red, early type galaxies, and becuase
of the red shift, it makes sense to use
new infra red bands rather than the
optical ones, as we go to higher[UNKNOWN].
This is fairly strightforward way but it's
very vulnerable to chance super-positions.
They can't sometimes tell whether it's a
cluster or it's just super position of
several smaller clumps along the line of
sight.
Probably the most famous of optical
cluster catalogs is the Abell Catalog,
which was initially compile by George
Abell from first Polymer sky surveys,
simply looking at the graphic plates with
a magnifier.
Later on, this was extended to the
southern sky as well, 4000 ga-, 4000
clusters.
This catalog has been a mainstay of
cluster studies for a long time.
Abell defined an operational criterion,
what is something if you will call a
cluster, there has to be certain number of
galaxies within certain magnitude range
and radius, and it's a reasonable set of
criteria, essentially a density contract
He divided the clusters in richness
classes, depending on how many galaxies he
could count in them.
He also had distance classes, because at
that time they didn't have directions to
all of them.
Abell's catalog had some problems.
I mean, it's not a statistical sample
although people often forget that.
Abell himself defined a small subset of it
that he thought would be.
Complete suitable for statistical studies
and everything else was extra.
So people complained about, subjective
nature of the catalog and these days
nobody's doing this by hand anymore.
We deploy algorithms that look for galaxy
or[UNKNOWN] using some.
Well defined criteria.
And typical modern catalogs of clusters of
galaxies using sky survey plates would
have 10s of 1,000s of clusters in.
The next method is using x-rays.
Clusters do contain large amounts of hot
gas, and they're very conspicuous on x-ray
sky.
Basically in the X-ray sky, you see point
sources which would be active nuclei or
maybe stars of some sort and extended
sources which are clusters of galaxies.
Here, the superposition problem does not
play a significant role.
In part, this is because the x-ray
emission is proportional to the square of
the density.
And that means you would have to have
material in one place to have a
significant brightness in x-rays.
But even so, x-ray method does not select
clusters by mass just like optical method
does not select them by mass.
Related to this is the Sunyaev-Zeldovich
effect, where we use combination of extra
measurements in micro background maps to
find clusters.
Now you may recall what Sunyaev-Zeldovich
effect is, we talked about this when we
talked about the distance scale.
If you look through a cloud of hot gas
towards the microwave background
photosphere, those photons sometimes
encounter a hot electron and get
universally come to a scattered, meaning
they get some of the electrons energy.
As a net result, the sppectrum of the
microbackground as seen through this cloud
of hot gas shifts to higher energies.
And now depending whether you're observing
it in a radio [inaudible] or whether the
peak of the black body you can either see
a bump or a dip in the sky.
And that bump or dip corresponds exactly
to the x-ray signal.
Finally, weak gravitational lensing
provides, at least in principle, means of
detecting clusters and this time by the
mass because that's what lensing is
actually sensitive to.
As you will recall, weak lensing will show
distorted images of galaxies due to some
foreground mass concentration such as
cluster, and, by looking for such signals
in the shape pattersn in background
galaxies, you could find Large masses
would or not.
They actually have any light or x-ray
emission.
As it turns out, to the best of my
knowledge, there was never a cluster
discovered using this method that wasn't
already seen or easily seen through
optical or x-ray.
So, what can you do with clusters?
We can measure some of their properties
more about those later, but you can just
look at them.
Typically in any imperical science, you
will begin with [unknown].
So, Abell initally classified clusters
simply on the basis of their appearance.
He noted that some are well relaxed
looking symmetric and the others still
look like raggedy.
Now we know that, that corresponds to
dynamically well formed versus still
forming clusters.
Now, in 1970s, Bautz and Morgan introduced
a different classification.
This one was not really based on
properties of cluster per se, but whether
or not it had a giant, dominant elliptical
galaxy in the middle as many of them do.
This turns out to have some significance
for galaxy evolution in clusters, but
again, it was relatively superficial.
Then further shifting focus, not on
cluster appearance, but on their galaxy
population, Oemler also classfied them
according to the Amont of spiral versus
elliptical galaxies in them.
That directly relates to the galaxy
evoltion processes in clusters.
And Rood and Sastry had one of these
schemes at which point the meager amount
of informaiton that's actually present in
these optical appearances was really
exhausted.
However, we do now have good physical
understanding where all this comes from.
Just a not about those central dominant
galaxies and clusters, or cD as they are
called.
They're giant ellipticals, but unlike
other giant ellipticals, they have an
extra fuzzy envelope at large radii.
Now we think that those extra fuzzy
envelopes do not really belong to the
galaxies, but to the clusters themselves.
That they are Piles of stars stripped away
by tidal interactions of galaxies in the
cluster.
They accumulate at the bottom of the
potential well and the B galaxy also
happens to be cospatial with them at the
bottom of the potential well.
So it looks as if it has this fuzzy
envelope.
So there's some interesting trends.
Well relaxed looking, symmetric round
clusters tend to be dominated by
elliptical galaxies.
Ruggedy, flat-ish looking clusters are
those where they're dominated by spiral
galaxies.
And even within given cluster, there is a
segregation, that the early type galaxies,
Elliptical,[INAUDIBLE] tend to congregate
in the inner portion of the cluster, where
spirals are predominately found at the
outskirts.
Now, we have good understanding
explanation effects from galaxy
revolution.
As far as the appearance is concerned, the
round, smooth ones, simply had enough time
to relax into varial or almost varial
distribution.
Whereas the lumpy, flattish, nonuniform
ones are still coming together.
And those Central dominant galaxies,
probably, are in part at least, product of
mergers of many smaller galaxies that
built them up.
So next time, we will talk about contents
of the clusters of galaxies.