So far, we have reviewed the morphology of
galaxies.
But now let's take a look at whether that
morphology actually correlates
meaningfully with some outer properties of
galaxies or their formation history.
First, let me critique the traditional
approach to galaxy morphology.
It is subjective, it is based on the
appearance, visual appearance in
particular wavelength, traditionally those
would be blue light sensitive photographic
plates.
But now we have the panchromatic view of
the universe, and when you look at
galaxies in different wavelength regimes
they look very different.
And, classification would presumably be
very different depending on which
wavelength regime was chosen.
Now we happened to have chosen visible
light and, in vicinity of visible light.
And you, if you look at bluer wavelengths,
ultraviolet included, you'll be very
sensitive to the star formation, so
galaxies will be clumpier, there'll be
more strong features due to the regions of
star formation.
If you look towards redder wavelengths,
which are not so sensitive to star
formation, but reflect the bulk of the
whole stellar population, galaxies will
look much smoother than that.
So there are three major problems with the
traditional morphological classification
of galaxies.
First it is subjective, it is based on
just the visual appearance, looking at
things and deciding whether some galaxies
have more open spiral arms or not.
However, these days we can actually,
process digital images of galaxies and
define in objective, consistent fashion
some of those structural parameters.
The second problem is that it's
superficial, it really is based on just
vis-, visual appearance of particular
wavelength region, and not on actual
physical properties that we'd be
interested in, such as galaxy mass for
example.
A more modern approach is to use
correlations, and clustering of different
galaxies and properties in parameter
spaces, and does objectively define galaxy
families using those correlations.
Finally it is an incomplete
classification, it was based on what was
known, more or less, in Hubble's days.
And it completely missed, a major
dichotomy between giant galaxies, by
giants I mean those that are usually seen
in Hubble sequence, And dwarfs, which turn
out to be largely a completely different
set of beasts, and I'll show you that
later.
And they themselves, may split in
different categories.
So that was completely missed by the
original, classification schemes.
Alright, so what does it mean?
Properties of galaxies, including their
appearance, are a product of their
formative and evolutionary histories.
And does, if we can interpret morphol,
morphological trends, then we can learn
something about galaxy formation and
galaxy evolution.
A lot can already be concluded from a very
basic fact.
Among the Hubble sequence galaxies at
least there seem to be two dominant
components.
The elliptical like bulge component and
disk component and the relative probab-,
presence of these two determines a lot of
properties of galaxies..
The bold slash elliptical stellar
components are older, and they're
schematically supported by random motions
of stars.
Stars have to somehow balance the
potential wells, in which they sit, and in
elliptical galaxies and bulges, those
motions are largely random, like molecules
in a gas, so we call them pressure.
Important.
In this galaxy, most of the kinetic energy
is in circular motion around the center.
And very productively little kinetic
energy is actually in random motion,
though there is some.
And so that's a major distinguishing
characteristic.
Note also that we'd be just talking about
light and we already know that the
dominant mass component is the dark
matter.
We think with some justification that dark
matter is probably also pressure
supported, random motions rather than
rotation.
And also, discs are certainly distinct
kinematical Age type components plus peril
arms which were used for original
classification, may be largely ornamental.
They are interesting patterns, but they do
not seem to really correlate very much
with anything else.
However, having said all that, there's
still very important and significant
trends along the Hubble sequence.
As we go from the early types ellipticals
and F zeros, towards later types of
spirals, F As, F Bs, F Cs, there are
several trends.
The age, average age of the population
decreases.
The disks are younger and the later Hubble
taps are even younger.
The start formation rate decreases.
There is almost none in elliptical and
bulges, and more and more as you go
towards the later types.
Because of that, the color also changes,
because young stellar populations are
dominated by luminous blue stars.
All stellar populations dominated by all
red giants, and so there would be turning
form redder colors towards the bluer
colors.
The gas content would change, at least the
neutral hydrogen gas will change.
There is almost none in ellipticals and
voltages except if it came from disks.
And there is more and more hydrogen
relative to the stellar mass as you go
towards later types.
Likewise, the outer components of these
cold interstellar medium, including dust,
are also increasing in their importance
towards the later types.
And finally, an important dynamical
characteristic which tells us a lot about
formative processes is that in the early
types, most of the kinetic energy is in
random motions.
In the later types, most of the kinetic
energy is in ordered, circular motions.
So because of these trends, and what they
really mean in direct interpretation,
Hubble classification persisted to this
day.
It's still useful, even though it's
superficial, incomplete and all that.
It's still fairly useful because it does
represent some important properties of
galaxies.
Not all, but, but some.
However it does not represent others, for
example, galaxy masses.
You can't think of any more fundamental
quantity than total mass of a galaxy that
does contain at all in Hubble
specifications.
In fact here are plots of the mean values
of important characteristics of galaxies
like radius, mass, luminosity and mass to
light ratios, as functions of the hubble
type.
As you can see these plots are largely
flat meaning there is independence of
these quantities on galaxies hubble type
except.
For some, maybe a little bit towards the
latest Hubble types.
If the Hubble type were representative or
indicative of, say, galaxy mass, there
should have been great correlations here
and there aren't any.
So this is really probably the most
fundamental problem with morphology based
classification as opposed to, say,
physical properties based classification.
All of these trends can be interpreted in
a simple way.
They are really a sequence of star
formation histories of galaxies.
Not just star formation today.
There is almost none in ellipticals and a
lot in spirals, and further away, go even
more, but integrated over the lifetimes.
Put simply, early types, ellipticals,
bulges, form most of their stars early on
and then very little after whereas disks
tend to have much more extended star
formation histories and it could be even
flat, uniform star formation through the
Hubble time.
For some of them the late type discs a
star formation may be even still
increasing in time.
Here is an interesting little fact.
Typical spirals say like Milky Way form
stars at typical rates of several solar
masses per year.
And if you add up all of the hydrogen
available for star formation, there is
usually about a billion solar masses or
so.
In other words if spirals were to continue
like this they'll burn through all of the
available hydrogen in a billion years or
less.
So it would seem as if we were really
living at special time.
More likely there is a fresh supply of
gas.
The fresh supply of gas that comes in from
the intergalactic medium and gets still
secreted from galaxies.
And here is, schematically, what I meant
by the trend in star formation histories.
If you plot, say, star formation rates as
a function of time in ellipticals and
bulges, there is a lot of activity early
on, dies off very quickly and there is
hardly any after that.
Whereas for these galaxies, there may be
little enhancement in the beginning, but
by and large remains flat.
Otherwise, I mentioned, it can even be
increasing.
So the simple picture can explain all of
the trends that we have seen so far.
Now let's recall the concept of stellar
populations as you probably remember from
earlier in astronomy study.
Those were introduced by Walter Baade who
noticed that there seemed to be two kinds
of stars.
There are younger stars that tends to be
constrict and to be in galaxy discs, and
older stars that tend to be in bulges or
galactic halos.
And he called them population 1 and
population 2.
So they differ in their ages and where
they're found, and sometimes in velocity,
and also their kinematics.
But as ideas can be extended, you can
think of stellar populations as
sub-systems inside galaxies.
They're characterized by their location,
by their dense distribution, by
kinematics, by the star formation history,
by the resulting metallicity, and so on.
And there's probably more than 2.
For example, in the milky way alone, we
count at least 4.
There would be old but metal rich bulge.
There'll be old but metal poor halo much
more extended, it'll be young stellar disk
where stars are now being formed and it'll
be older and somewhat thicker disk
composed of older stars.
Again note dark matter is yet whole
another issue here.
How can we understand the connection
between star formation history and
dynamics?
Well, schematically, it works like this.
Stars are essentially mass points, and
even in colliding galaxies, galaxy
mergers, they are behaving like
dissipation systems.
They just follow what ever potential there
is.
And they, in a sense, dynamically remember
the dynamics of their birth.
So if you start with a bunch of small
galaxies merging together in a random
fashion, then stars that used to belong to
them will continue to move in random
directions, and you'll end up with a
stellar system that is supported by random
motions which is just like elliptical
galaxies.
Now, consider collapse, not of galaxies
already made of some stars, but just
hydrogen clouds.
They dissipate energy, but they cannot
dissipate angular momentum, therefore,
they settle.
Configuration that gives them minimum
energy for the given amount of angular
momentum which is a flat, thin rotating
disk.
Now they make stars and those stars then
remember dynamics of their birth and they
continue moving in circular orbits.
There is very little in terms of random
motions.
So in this way we can connect the dynamics
of stellar populations or subsystem of
these galaxies with the histories of their
star formation.
What about their metallicities?
Remember that enrichment of stellar,
interstellar medium comes from stars
themselves.
They form stars, massive ones explode,
disseminate heavier elements, they cooked
up in interstellar medium, new stars are
formed from that, and so on.
Now if you have a really massive big
galaxy, supernova ejecta will not be able
to escape the potential well, and they
will mix in with the rest of the gas,
which serves as a fuel for the next
generation of stars and the metallicity
will gradually increase in time.
On the other hand, if you have a very low
mass host system, a little dwarf galaxy or
protogalactic fragment, the potential low
is not so deep and the kinetic energy of
supernova ejecta is sufficient to expel
them out into the intergalactic space.
So that little galaxy does not evolve
chemically very much.
It does a little bit but its metallicity
remains low and doesn't change very much
in time.
So that's the qualitative picture but
let's try to put this in a quantitative
footing.
We would like to know about actual
distributions and correlations of
meaningful physical properties of galaxies
that we can measure in well defined
fashion, and those include first the
density distribution.
How are they distributed spatially?
What is the density profile?
Their kinematical profile, how is the
kinematics changing as a function of
radius?
The relative importance of Auld and Yang,
or rotating in random components, and the
chemical abundances and how they change.
So by and large, structural properties or
density distribution are obtained through
photometry , or surface photometry which
is spacial result of pretty much
everything else, kinematics,
metallicities, star formation rates, comes
from spectroscopic measurements.
And in order to interpret these
measurements, we have to use various
population synthesis models and dynamical
models.
Dynamical models can be assumed as toy
models with some gravitational potential.
See where that goes.
Stellar population synthesis models,
meaning what Kind of mix of different
stars would be required to obtain spectrum
as observed, and of course because stars
evolve in time, their mixture will evolve
and the spectrum will evolve in time and
so we need galaxy evolution models like
that.
We have all of those and we'll talk about
them a little later.
So next we will turn to the structural
properties of spiral galaxies.