# What are the criteria for the classification of galaxies?

We are searching data for your request:

Forums and discussions:
Manuals and reference books:
Data from registers:
Wait the end of the search in all databases.
Upon completion, a link will appear to access the found materials.

There are lots of galaxies, for example, the Milky Way and so on. These galaxies consist of lots of stars. I want to know how galaxies are classified. Is it only by their shape that is caused by gravity? Or are there some other criteria?

A galaxy is just a very large collection of stars (and interstellar matter such as dark matter, gas and dust) held together by gravity. Galaxies are classified mainly as elliptical, spiral, and irregular. There is not a specific criterion other than the fact that the stars all all bound by their mutual gravitational attraction. It is speculated that most galaxies have a black hole at its center. This is not necessary however for a collection of stars to be considered as a galaxy.

There are three main classes of galaxies: Irregulars, Ellipticals, and Spirals. Irregular galaxies, as their name suggests, do not fit into the "normal" classification scheme.

So, how do we distinguish between elliptical and spiral galaxies?

Brighness profile

The radial brightness profile of an elliptical galaxy follows a deVaucouleur law ($r^{1/4}$).

Spiral galaxies have an exponential radial brightness profile, although their central regions ("bulge") also follows a deVaucouleur law.

Star formation

Stars are formed in the spiral arms of spiral galaxies (and can be formed in irregulars), while elliptical galaxies tend to only have old, and consequently low mass, stars.

Components

As far as we can tell, all galaxies consist of a dark matter halo and stars. In addition, spiral galaxies also have clouds of dust and gas. If conditions are right, these can form new stars. (Some ellpticals have a very thin, very hot gas component as well, but there is a lot less of it than in a spiral galaxy).

Kinematics

Spiral galaxies are rotationally supported, while elliptical galaxies are mainly pressure-supported (i.e. they act like an ideal gas, with stars as gas molecules). There are some rotational features present in ellipticals, but they tend to be minor compared to the overall random motion.

A graphical overview of the various galaxy types is usually shown in the Hubble tuning fork diagram. Note that this does not indicate an evolutionary progression from one type to the next.

Alex answers nicely how galaxies can be classified according to their morphology. In astronomy, galaxies are detected using a variety of detection techniques. Especially in the high-redshift (i.e. distant) Universe, galaxies are not easily detected and are only visible using specific methods (although some galaxies show up with multiple techniques). These methods each probe different things, and galaxies belonging to one class will thus have other parameters than galaxies belong to other classes, although there will always be ome overlap.

The physical properties defining whether a galaxy may be selected by a given technique is hence not only morphology, but also stellar mass, star formation rate, dust mass, size, clumpiness, kinematics, luminosity, the presence of active galactic nuclei, and many others.

Accordingly, we ofted classify galaxies from the method and the selection criterion used (and preferably use a three-letter acronym to describe them):

Some of these are:

## Lyman-break galaxies (LBGs)

The technique used to find these galaxies revolutionized the field in the mid-90'es (Steidel et al. 1996). The reason is that a large field of view can be investigated, allowing to detect many galaxies at the same time. The idea is to observe the same field in several different wavelength bands. If large amounts of neutral hydrogen is present, wavelengths shortward of the "Lyman-break" at 912 Å, or 91.2 nm, needed to ionize hydrogen are absorbed, effectively making the galaxy invisible in all bands shortward of this. And because the light is redshifted as it travel through the Universe toward us, galaxies at different redshifts will drop out of different band (the method is also called the "drop-out technique"). An example is seen here, where the galaxy spectrum (black line) drops steeply so that flux enters the red $R$ band and the green $G$ band, but not the ultraviolet $U$ band:

Credit: J. Fynbo.

In the above figure, the break has been redshifted to somewhere between the $G$ band and the $U$ band, constraining its redshift to roughly $z = 3$-$4$. To further constrain the redshift, spectroscopic follow-up is needed.

Since large amounts of neutral hydrogen is needed, this technique will tend to select massive, and hence rather evolved galaxies.

## Lyman alpha emitters (LAEs)

When an electron decays from the first excited state to the ground state, a s-called Lyman alpha (Ly$alpha$) photon is emitted. This can happen either when a hydrogen atom is perturbed (in a collision) and excited, or if hydrogen is ionised and recombines. Both mechanisms are at play when galaxies are born, where gas accretes onto a central potential (causing collisions), and young, massive star ionize the surrounding gas.

Galaxies found from their ability to emit Ly$alpha$ are called LAEs. They can be either found either spectroscopically - where are strong emission line will be seen at $lambda = 1216$ Å - or photometrically by observing the field in a broadband and a narrowband centered at $lambda = 1216$ Å and looking for excess flux in the narrowband.

Since this techniques tends to probe young galaxies, they will often be relatively small, but with a high star formation rate. And because dust absorbs Ly$alpha$ more easily than other wavelengths, LAEs tend to be rather dust-free.

## Sub-millimeter galaxies (SMGs)

If a galaxy, on the other hand, is very dusty, it can be difficult to detect in the optical and, especially, in the ultraviolet. The reason is that dust has a strong preference to absorb light with shorter wavelengths. But the energy absorbed must go somewhere, and is thus emitted again, although at longer wavelengths, i.e. in the infrared and in the sub-mm region. Galaxies found this way are referred to as SMGs. Because it takes some time for the dust mass to build up, this technique tends to probe evolved, massive galaxies.

## Damped Lyman alpha absorbers (DLAs)

The three techniques described above all have in common that they detect galaxies from their emission. A complimentary technique is looking for absorption features in the spectrum of a bright background source, e.g. a quasar. Because the light is redshifted on its way, especially hydrogen but also metals such as iron and magnesium produce absorption lines at various places in the spectrum corresponding to the wavelength that the quasar light has been redshifted to at a given point in space. Diffuse hydrogen filaments make narrow absorption line known as the Lyman $alpha$ forest, and when a large pocket of gas is present - which indicates the presence of a galaxy - a broad ("damped") absorption line is produced.

An example is seen in this spectrum of the quasar Q2348-011 lying at $z=3.0$. An intervening galaxy at $z=2.6$ causes the broad absorption at $lambdasim4400$ Å.

Credit: Laursen (2010).

Only in few cases is the galaxy responsible for the absorption found. This is partly because the light from the quasar outshines everything in its (projected) vicinity, but possibly also because the huge hydrogen cloud is a galaxy in the making, that perhaps hasn't form many stars yet. And since the probabiliy for sightlines toward quasars of hitting a small galaxy is larger than hitting a large galaxy (due to the total cross section of small galaxies being larger), galacitc counterparts of DLAs should tend to be small. Thus, DLAs are thought to probe young galaxies in the process of forming.

## Other types

Other types includes distant red galaxies (DRGs), (ultra)luminous infrared galaxies (LIRGs and ULIRGs), and gamma-ray burst host galaxies (GHGs). One of the big challenges of astronomy is to ascertain how the galaxies of the different group fit in some big picture. Is there for instance an evolutionary sequence from DLA→LAE→LBG→SMG→DRG (see e.g. Gawiser 2005)?

The primary, traditional classification for galaxies in the local universe is based on "morphology" -- in other words, on their optically visible shape; this goes back to the Hubble Sequence.

I'll list the main categories and the defining shape, and then some other characteristics which are not part of the main criteria.

Elliptical Galaxies: These are circular or elliptical in projected shape (ellipsoidal or triaxial in 3D shape), with no visible disk and very little gas or dust, and little or no evidence for young stars.

The stars are almost all old, and tend to orbit in random directions. Very luminous/massive ellipticals tend to have centrally concentrated radial profiles in the stellar density (now usually described using Sersic profiles with high values of the index $n$); faint, low-mass "dwarf ellipticals" have more exponential stellar profiles.

S0 (or Lenticular) Galaxies: These have a prominent disk of stars, but one which lacks visible spiral arms and has little or no gas or dust, and little or no evidence for young stars. The disk may, however, have one (or sometimes two) stellar bars, and sometimes rings as well.

The stars are mostly old and almost all orbit in the same direction within the disk, but the orbits may be somewhat elliptical rather than circular. They almost always have a prominent "bulge" of stars dominating the middle of the galaxy; the bulge may be a very centrally concentrated part of the disk, the vertically thickened part of a stellar bar, or a round collection of old stars with mostly random orbits (somewhat like a small elliptical galaxy) -- or a combination of all three.

Spiral Galaxies: These have a prominent disk of stars, gas and dust; the disk has spiral arms in it (hence the name). The subclassifications within this category (e.g., Sa vs Sb vs Sc vs Sd) are based on a combination of three factors: the relative prominence of a central bulge (if any); how tightly or loosely wound the spiral arms appear to be; and the degree to which the spiral arms are smooth versus being broken up into fragments and stellar clusters.

The stars and gas almost all rotate in the same direction, with orbits that are relatively circular. They are almost always a mix of young and old stars, with new stars being formed in the disk. They may have a bulge in the center, but some do not; the bulges may be as diverse and complicated as those in S0 galaxies.

Irregular Galaxies: As the name suggests, these are more raggedy, lopsided, and generally "shapeless". They are usually rich in gas, and are almost always lower in mass than the other types; they are, like spirals, often forming stars at the present time.

There are a number of different kind of dwarf (= faint, low-mass) galaxies which may or may not fall neatly into the above categories. For example, dwarf spheroidal galaxies are very faint and low-mass; in terms of structure, stellar orbits, and the absence of gas or current star formation, they resemble ellipticals, but are very diffuse rather than centrally concentrated. Recent and still somewhat mysterious discoveries include "ultracompact dwarf" (UCD) galaxies and "ultradiffuse galaxies".

## The Classification of Galaxies

A useful first step towards an understanding of galaxies is a classification based on their various forms. Although such a morphological classification must always be to some extent subjective, it provides a framework within which the quantitative properties of galaxies can be discussed in a systematic fashion. However, it should always be remembered that the picture thus obtained will be limited to those galaxies that are large and bright enough to be easily visible in the sky. An idea of the consequent limitations can be obtained from Fig. 18.1, showing the radii and magnitudes of normal galaxies. One sees that only within a narrow region of this diagram can galaxies be easily found. If a galaxy has too large a radius for its magnitude (small surface brightness), it will disappear in the background light from the

Fig. 18.1. Magnitudes and diameters of observable extragalac-tic objects. Objects to the upper left look like stars. The quasars in this region have been discovered on the basis of their spectra. Objects to the lower right have a surface brightness much smaller than that of the night sky. In recent years large numbers of low surface brightness galaxies have been discovered in this region. (Arp, H. (1965): Astrophys. J. 142, 402)

Hannu Karttunen et al. (Eds.), Galaxies.

In: Hannu Karttunen et al. (Eds.), Fundamental Astronomy, 5th Edition. pp. 367-391 (2007) DOI: 11685739_18 © Springer-Verlag Berlin Heidelberg 2007

night sky. On the other hand, if its radius is too small, it looks like a star and is not noticed on a photographic plate. In the following, we shall mainly be concerned with bright galaxies that fit within these limits.

If a classification is to be useful, it should at least roughly correspond to important physical properties of the galaxies. Most classifications accord in their main features with the one put forward by Edwin Hubble in 1926. Hubble's own version of the Hubble sequence is shown in Fig. 18.2. The various types of galaxies are ordered in a sequence from early to late types. There are three main types: elliptical, lenticular, and spiral galaxies. The spirals are divided into two sequences, normal and barred spirals. In addition, Hubble included a class of irregular galaxies.

The elliptical galaxies appear in the sky as elliptical concentrations of stars, in which the density falls off in a regular fashion as one goes outwards. Usually there are no signs of interstellar matter (dark bands of dust, bright young stars). The ellipticals differ from each other only in shape and on this basis they are classified as E0, E1. E7. If the major and minor axes of an elliptical galaxy are a and b, its type is defined to be En, where n = 10[ 1 - -

An E0 galaxy thus looks circular in the sky. The apparent shape of an E galaxy depends on the direction from which it is seen. In reality an E0 galaxy may therefore be truly spherical or it may be a circular disc viewed directly from above.

A later addition to the Hubble sequence is a class of giant elliptical galaxies denoted cD. These are gener ally found in the middle of clusters of galaxies. They consist of a central part looking like a normal elliptical surrounded by an extended fainter halo of stars.

In the Hubble sequence the lenticulars or S0 galaxies are placed between the elliptical and the spiral types. Like the ellipticals they contain only little interstellar matter and show no signs of spiral structure. However, in addition to the usual elliptical stellar component, they also contain a flat disc made up of stars. In this respect they are like spiral galaxies (Figs. 18.3, 18.4).

The characteristic feature of spiral galaxies is a more or less well-defined spiral pattern in the disc. Spiral galaxies consist of a central bulge, which is structurally similar to an E galaxy, and of a stellar disc, like in an S0 galaxy. In addition to these, there is a thin disc of gas and other interstellar matter, where young stars are being born, forming the spiral pattern. There are two sequences of spirals, normal Sa-Sb-Sc, and barred SBa-SBb-SBc spirals. In the barred spirals the spiral pattern ends at a central bar, whereas in the normal spirals the spiral pattern may end at an inner ring or continue all the way to the centre. The position of a galaxy within the spiral sequence is determined on the basis of three criteria (which are not always in agreement): later types have a smaller central bulge, more narrow spiral arms and a more open spiral pattern. The Milky Way Galaxy is thought to be of type SABbc (intermediate between Sb and Sc, and between normal and barred spirals).

The classical Hubble sequence is essentially based on bright galaxies faint galaxies have been less easy to fit into it (Fig. 18.5). For example, the irregular galaxies of the original Hubble sequence can be divided into the classes Irr I and Irr II. The Irr I galaxies form a continua-

Fig. 18.2. The Hubble sequence in Hubble's 1936 version. At this stage the existence of type S0 was still doubtful. Photographs of the Hubble types are shown in Figs. 18.6 and 18.15 (E) 18.3 and 18.4 (S0 and S) 18.12 (S and Irr II) 18.5 (Irr I and dE). (Hubble, E.P. (1936): The Realm of the Nebulae (Yale University Press, New Haven))

Fig. 18.2. The Hubble sequence in Hubble's 1936 version. At this stage the existence of type S0 was still doubtful. Photographs of the Hubble types are shown in Figs. 18.6 and 18.15 (E) 18.3 and 18.4 (S0 and S) 18.12 (S and Irr II) 18.5 (Irr I and dE). (Hubble, E.P. (1936): The Realm of the Nebulae (Yale University Press, New Haven))

NGC488 Type Sab NGC628 M74 TypeSc

Fig. 18.3. The classification of normal spiral and S0 galaxies. (Mt. Wilson Observatory)

NGC488 Type Sab NGC628 M74 TypeSc

Fig. 18.3. The classification of normal spiral and S0 galaxies. (Mt. Wilson Observatory)

Type SBc(sr)

Fig. 18.4. Different types of SB0 and SB galaxies. The type (r) or (s) depends on whether the galaxy has a central ring or not. (Mt. Wilson Observatory)

Fig. 18.5. Above: The Small Magellanic Cloud (Hubble type Irr I), a dwarf companion of the Milky Way. (Royal Observatory, Edinburgh). Below: The Sculptor Galaxy, a dE dwarf spheroidal. (ESO)

Fig. 18.6. M32 (type E2), a small elliptical companion of the Andromeda Galaxy. (NOAO/Kitt Peak National Observatory)

Fig. 18.6. M32 (type E2), a small elliptical companion of the Andromeda Galaxy. (NOAO/Kitt Peak National Observatory)

tion of the Hubble sequence towards later types beyond the Sc galaxies. They are rich in gas and contain many young stars. Type Irr II are dusty, somewhat irregular small ellipticals. Other types of dwarf galaxies are often introduced. One example is the dwarf spheroidal type dE, similar to the ellipticals, but with a much less centrally concentrated star distribution. Another is the blue compact galaxies (also called extragalactic HII regions), in which essentially all the light comes from a small region of bright, newly formed stars.

### 18.2 Luminosities and Masses

Distances. In order to determine the absolute luminosities and linear dimensions of galaxies one needs to know their distances. Distances are also needed in order to estimate the masses of galaxies, because these estimates depend on the absolute linear size. Distances within the Local Group can be measured by the same methods as inside the Milky Way, most importantly by means of variable stars. On the very large scale (beyond 50 Mpc), the distances can be deduced on the basis of the expansion of the Universe (see Sect. 19.1). In order to connect these two regions one needs methods of distance determination based on the properties of individual galaxies.

To some extent local distances can be determined using structural components of galaxies, such as the sizes of H II regions or the magnitudes of globular clusters. However, to measure distances of tens of megaparsecs, one needs a distance-independent method to determine the absolute luminosities of entire galaxies. Several such methods have been proposed. For example, a luminosity classification has been introduced for late spiral types by Sidney van den Bergh. This is based on a correlation between the luminosity of a galaxy and the prominence of its spiral pattern.

Other distance indicators are obtained if there is some intrinsic property of the galaxy, which is correlated with its total luminosity, and which can be measured independently of the distance. Such properties are the colour, the surface brightness and the internal velocities in galaxies. All of these have been used to measure distances to both spiral and elliptical galaxies. For example, the absolute luminosity of a galaxy should depend on its

Fig. 18.7. Compound luminosity function of thirteen clusters of galaxies. The open symbols have been obtained by omitting the cD galaxies. The distribution is then well described by (18.2). The cD galaxies (filled symbols) cause a deviation at the bright end. (Schechter, P. (1976): Astrophys. J. 203, 297)

measure the luminosity of a galaxy out to a given value of the surface brightness, e.g. to 26.5mag/sq.arcsec. For a given Hubble type, the total luminosity L may vary widely.

As in the case of stars, the distribution of galaxy luminosities is described by the luminosity function 0( L). This is defined so that the space density of galaxies with luminosities between L and L + d L is 0( L) d L .It can be determined from the observed magnitudes of galaxies, once their distances have been estimated in some way. In practice, one assumes some suitable functional form for 0(L), which is then fitted to the observations. One common form is Schechter's luminosity function,

Fig. 18.7. Compound luminosity function of thirteen clusters of galaxies. The open symbols have been obtained by omitting the cD galaxies. The distribution is then well described by (18.2). The cD galaxies (filled symbols) cause a deviation at the bright end. (Schechter, P. (1976): Astrophys. J. 203, 297)

mass. The mass, in turn, will be reflected in the velocities of stars and gas in the galaxy. Accordingly there is a relationship between the absolute luminosity and the velocity dispersion (in ellipticals) and the rotational velocity (in spirals). Since rotational velocities can be measured very accurately from the width of the hydrogen 21-cm line, the latter relationship (known as the Tully-Fisher relation) is perhaps the best distance indicator currently available.

The luminosity of the brightest galaxies in clusters has been found to be reasonably constant. This fact can be used to measure even larger distances, providing a method which is important in cosmology.

Luminosities. The definition of the total luminosity of a galaxy is to some extent arbitrary, since galaxies do not have a sharp outer edge. The usual convention is to

The values of the parameters 0*, L*, a are observation-ally determined for different types of objects in general, they will be functions of position.

The shape of the luminosity function is described by the parameters a and L *. The relative number of faint galaxies is described by a. Since its observed value is about —1.1, the density of galaxies grows monotonically as one goes towards fainter luminosities. The luminosity function falls off steeply above the luminosity L *, which therefore represents a characteristic luminosity of bright galaxies. The observed L* corresponds to an absolute magnitude M* = — 21.0 mag. The corresponding magnitude for the Milky Way Galaxy is probably — 20.2 mag. The cD giant galaxies do not obey this brightness distribution their magnitudes may be —24 mag and even brighter.

The parameter 0* is proportional to the space density of galaxies and is therefore a strong function of position. Since the total number density of galaxies predicted by relation (18.2) is infinite, we define n* = density of galaxies with luminosity > L *. The observed average value of n* over a large volume of space is n* = 3.5 x 10—3 Mpc—3. The mean separation between galaxies corresponding to this density is 4 Mpc. Since most galaxies are fainter than L *, and since, in addition, they often belong to groups, we see that the distances between normal galaxies are generally not much larger than their diameters.

Masses. The distribution of mass in galaxies is a crucial quantity, both for cosmology and for theories of the origin and evolution of galaxies. Observationally it is determined from the velocities of the stars and interstellar gas. Total masses of galaxies can also be derived from their motions in clusters of galaxies. The results are usually given in terms of the corresponding mass-luminosity ratio M/L, using the solar mass and luminosity as units. The value measured in the solar neighbourhood of the Milky Way is M/L = 3. If M/ L were constant, the mass distribution could be determined from the observed luminosity distribution by multiplying with M/L.

The masses of eliptical galaxies may be obtained from the stellar velocity dispersion given by the broadening of spectral lines. The method is based on the virial theorem (see Sect. 6.10), which says that in a system in equilibrium, the kinetic energy T and the potential energy U are related according to the equation

Since ellipticals rotate slowly, the kinetic energy of the stars may be written

where M is the total mass of the galaxy and v the velocity width of the spectral lines. The potential energy is

where R is a suitable average radius of the galaxy that can be estimated or calculated from the light distribution. Introducing (18.4) and (18.5) into (18.3) we obtain:

From this formula the mass of an elliptical galaxy can be calculated when v2 and R are known. Some observations of velocities in elliptical galaxies are given in Fig. 18.8. These will be further discussed in Sect. 18.4. The value of M/L derived from such observations is about 10 within a radius of 10 kpc. The mass of a bright elliptical might thus be up to 1013 M0.

The masses of spiral galaxies are obtained from their rotation curve v(R), which gives the variation of their rotational velocity with radius. Assuming that most of the mass is in the almost spherical bulge, the mass within radius R, M(R), can be estimated from Kepler's third law:

Some typical rotation curves are shown in Fig. 18.9. In the outer parts of many spirals, v(R) does not depend on R. This means that M(R) is directly proportional to the radius - the further out one goes, the larger the interior mass is. Since the outer parts of spirals are very faint, at large radii the value of M/L is directly proportional to the radius. For the disc, one finds that M/L = 8 for early and M/L = 4 for late spiral types. The largest measured total mass is 2 x 1012 M0.

Fig. 18.8. Velocity of rotation V(R) [km s 1 ] and velocity dispersion a( R) [kms-1] as functions of radius [kpc] for types E2

Fig. 18.8. Velocity of rotation V(R) [km s 1 ] and velocity dispersion a( R) [kms-1] as functions of radius [kpc] for types E2

and E5. The latter galaxy is rotating, the former is not. (Davies, R. L. (1981): Mon. Not. R. Astron. Soc. 194, 879)

and E5. The latter galaxy is rotating, the former is not. (Davies, R. L. (1981): Mon. Not. R. Astron. Soc. 194, 879)

Fig. 18.9. Rotation curves for seven spiral galaxies. (Rubin, V.C., Ford, W.K., Thonnard, N. (1978): As-trophys. J. (Lett.) 225, L107)

In order to measure the mass at even larger radii where no emission can be detected, motions in systems of galaxies have to be used. One possibility is to use pairs of galaxies. In principle, the method is the same as for binary stars. However, because the orbital period of a binary galaxy is about 109 years, only statistical information can be obtained in this way. The results are still uncertain, but seem to indicate values of M/L = 20-30 at pair separations of about 50 kpc.

A fourth method to determine galaxy masses is to apply the virial theorem to clusters of galaxies, assuming that these are in equilibrium. The kinetic energy T in (18.4) can then be calculated from the observed red-shifts and the potential energy U , from the separations between cluster galaxies. If it is assumed that the masses of galaxies are proportional to their luminosities, it is found that M/L is about 200 within 1 Mpc of the cluster centre. However, there is a large variation from cluster to cluster.

Present results suggest that as one samples larger volumes of space, one obtains larger values for the mass-luminosity ratio. Thus a large fraction of the total mass of galaxies must be in an invisible and unknown form, mostly found in the outer parts. This is known as the missing mass problem, and is one of the central unsolved questions of extragalactic astronomy.

### 18.3 Galactic Structures

Ellipticals and Bulges. In all galaxies the oldest stars have a more or less round distribution. In the Milky Way this component is represented by the population II stars. Its inner parts are called the bulge, and its outer parts are often referred to as the halo. There does not appear to be any physically significant difference between the bulge and the halo. The population of old stars can be best studied in ellipticals, which only contain this component. The bulges of spiral and S0 galaxies are very similar to ellipticals of the same size.

The surface brightness distribution in elliptical galaxies essentially depends only on the distance from the centre and the orientation of the major and minor axis. If r is the radius along the major axis, the surface brightness I(r) is well described by de Vaucouleurs' law:

The constants in (18.8) have been chosen so that half of the total light of the galaxy is radiated from within the radius re and the surface brightness at that radius is Ie. The parameters re and Ie are determined by fitting (18.8) to observed brightness profiles. Typical values for r e

elliptical, normal spiral and SO galaxies are in the ranges re = 1-10 kpc and Ie corresponds to 20-23 magnitudes per square arc second.

Although de Vaucouleurs' law is a purely empirical relation, it still gives a remarkably good representation of the observed light distribution. However, in the outer regions of elliptical galaxies, departures may often occur: the surface brightness of dwarf spheroidals often falls off more rapidly than (18.8), perhaps because the outer parts of these galaxies have been torn off in tidal encounters with other galaxies. In the giant galaxies of type cD, the surface brightness falls off more slowly (see Fig. 18.10). It is thought that this is connected with their central position in clusters of galaxies.

Although the isophotes in elliptical galaxies are ellipses to a good approximation, their ellipticities and the orientation of their major axes may vary as a function of radius. Different galaxies differ widely in this respect, indicating that the structure of ellipticals is not as simple as it might appear. In particular, the fact that the direction of the major axis sometimes changes within a galaxy suggests that some ellipticals may not be axially symmetric in shape.

From the distribution of surface brightness, the three-dimensional structure of a galaxy may be inferred as explained in *Three-Dimensional Shape of Galaxies.

The relation (18.8) gives a brightness profile which is very strongly peaked towards the centre. The real distribution of axial ratios for ellipticals can be statistically inferred from the observed one. On the (questionable) assumption that they are rotationally symmetric, one obtains a broad distribution with a maximum corresponding to types E3-E4. If the true shape is not axisymmetric, it cannot even statistically be uniquely determined from the observations.

Discs. A bright, massive stellar disc is characteristic for S0 and spiral galaxies, which are therefore called disc galaxies. There are indications that in some ellipticals there is also a faint disc hidden behind the bright bulge. In the Milky Way the disc is formed by population I stars.

The distribution of surface brightness in the disc is described by the expression

Figure 18.11 shows how the observed radial brightness distribution can be decomposed into a sum of two components: a centrally dominant bulge and a disc contributing significantly at larger radii. The central surface brightness I0 typically corresponds to 21-22 mag./sq.arcsec, and the radial scale

Fig. 18.10. The distribution of surface brightness in E and cD galaxies. Ordinate: surface magnitude, mag/sq.arcsec abscissa: (radius [kpc])1/4. Equation (18.8) corresponds to a straight line in this representation. It fits well with an E galaxy, but for type cD the luminosity falls off more slowly in the outer regions. Comparison with Fig. 18.11 shows that the brightness distribution in S0 galaxies behaves in a similar fashion. cD galaxies have often been erroneously classified as S0. (Thuan, T.X., Romanishin, W. (1981): Astrophys. J. 248, 439)

Figure 18.11 shows how the observed radial brightness distribution can be decomposed into a sum of two components: a centrally dominant bulge and a disc contributing significantly at larger radii. The central surface brightness I0 typically corresponds to 21-22 mag./sq.arcsec, and the radial scale

Fig. 18.11. The distribution of surface brightness in types SO and Sb. Ordinate: mag/sq.arc sec abscissa: radius [arc sec]. The observed surface brightness has been decomposed into a sum of bulge and disc contributions. Note the larger disc component in type Sb. (Boroson, T. (1981): As-trophys. J. Suppl. 46, 177)

Fig. 18.11. The distribution of surface brightness in types SO and Sb. Ordinate: mag/sq.arc sec abscissa: radius [arc sec]. The observed surface brightness has been decomposed into a sum of bulge and disc contributions. Note the larger disc component in type Sb. (Boroson, T. (1981): As-trophys. J. Suppl. 46, 177)

length r0 = 1-5 kpc. In Sc galaxies the total brightness of the bulge is generally slightly smaller than that of the disc, whereas in earlier Hubble types the bulge has a larger total brightness. The thickness of the disc, measured in galaxies that are seen edge-on, may typically be about 1.2 kpc. Sometimes the disc has a sharp outer edge at about 4 r0.

The Interstellar Medium. Elliptical and SO galaxies contain very little interstellar gas. However, in some ellipticals neutral hydrogen amounting to about 0.1% of the total mass has been detected, and in the same galaxies there are also often signs of recent star formation. In some S0 galaxies much larger gas masses have been observed, but the relative amount of gas is very variable from one galaxy to another. The lack of gas in these galaxies is rather unexpected, since during their evolution the stars release much more gas than is observed.

The relative amount of neutral hydrogen in spiral galaxies is correlated with their Hubble type. Thus Sa spirals contain about 2%, Sc spirals 10%, and Irr I galaxies up to 30% or more.

The distribution of neutral atomic hydrogen has been mapped in detail in nearby galaxies by means of radio observations. In the inner parts of galaxies the gas forms a thin disc with a fairly constant thickness of about 200 pc, sometimes with a central hole of a few kpc diameter. The gas disc may continue far outside the optical disc, becoming thicker and often warped from the central disc plane.

Most of the interstellar gas in spiral galaxies is in the form of molecular hydrogen. The hydrogen molecule cannot be observed directly, but the distribution of carbon monoxide has been mapped by radio observations. The distribution of molecular hydrogen can then be derived by assuming that the ratio between the densities of CO and H2 is everywhere the same, although this may not always be true. It is found that the distribution obeys a similar exponential law as the young stars and HII regions, although in some galaxies (such as the Milky Way) there is a central density minimum. The surface density of molecular gas may be five times larger than that of H I, but because of its strong central concentration its total mass is only perhaps two times larger.

The distribution of cosmic rays and magnetic fields in galaxies can be mapped by means of radio observations of the synchrotron radiation from relativistic electrons. The strength of the magnetic field deduced in this way is typically 0.5-1 nT. The observed emission is polarized, showing that the magnetic field is fairly well-ordered on large scales. Since the plane of polarization is perpendicular to the magnetic field, the large-scale structure of the magnetic field can be mapped. However, the plane of polarization is changed by Faraday rotation, and for this reason observations at several wavelengths are needed in order to determine the direction of the field. The results show that the field is generally strongest in the plane of the disc, and is directed along the spiral arms in the plane. The field is thought to have been produced by the combined action of rising elements of gas, perhaps

produced by supernova explosions, and the differential rotation, in principle in the same way as the production of solar magnetic fields was explained in Chapter 12.

* Three-Dimensional Shape of Galaxies

Equations (18.8) and (18.9) describe the distribution of galactic light projected on the plane of the sky. The actual three-dimensional luminosity distribution in a galaxy is obtained by inverting the projection. This is easiest for spherical galaxies.

Let us suppose that a spherical galaxy has the projected luminosity distribution I(r) (e.g. as in (18.8)). With coordinates chosen according to the figure, I(r) is given in terms of the three-dimensional luminosity distribution p(R) by

This is known as an Abel integral equation for p(R), and has the solution p( R) = -

## Spiral Galaxies

Our own Galaxy and the Andromeda galaxy are typical, large spiral galaxies. They consist of a central bulge, a halo, a disk, and spiral arms. Interstellar material is usually spread throughout the disks of spiral galaxies. Bright emission nebulae and hot, young stars are present, especially in the spiral arms, showing that new star formation is still occurring. The disks are often dusty, which is especially noticeable in those systems that we view almost edge on (Figure).

Figure 1: Spiral Galaxies. (a) The spiral arms of M100, shown here, are bluer than the rest of the galaxy, indicating young, high-mass stars and star-forming regions. (b) We view this spiral galaxy, NGC 4565, almost exactly edge on, and from this angle, we can see the dust in the plane of the galaxy it appears dark because it absorbs the light from the stars in the galaxy. (credit a: modification of work by Hubble Legacy Archive, NASA, ESA, and Judy Schmidt credit b: modification of work by “Jschulman555″/ Wikimedia)

In galaxies that we see face on, the bright stars and emission nebulae make the arms of spirals stand out like those of a pinwheel on the fourth of July. Open star clusters can be seen in the arms of nearer spirals, and globular clusters are often visible in their halos. Spiral galaxies contain a mixture of young and old stars, just as the Milky Way does. All spirals rotate, and the direction of their spin is such that the arms appear to trail much like the wake of a boat.

About two-thirds of the nearby spiral galaxies have boxy or peanut-shaped bars of stars running through their centers (Figure 2). Showing great originality, astronomers call these galaxies barred spirals.

Figure 2: Barred Spiral Galaxy. NGC 1300, shown here, is a barred spiral galaxy. Note that the spiral arms begin at the ends of the bar. (credit: NASA, ESA, and the Hubble Heritage Team(STScI/AURA))

As we noted in The Milky Way Galaxy chapter, our Galaxy has a modest bar too. The spiral arms usually begin from the ends of the bar. The fact that bars are so common suggests that they are long lived it may be that most spiral galaxies form a bar at some point during their evolution.

In both barred and unbarred spiral galaxies, we observe a range of different shapes. At one extreme, the central bulge is large and luminous, the arms are faint and tightly coiled, and bright emission nebulae and supergiant stars are inconspicuous. Hubble, who developed a system of classifying galaxies by shape, gave these galaxies the designation Sa. Galaxies at this extreme may have no clear spiral arm structure, resulting in a lens-like appearance (they are sometimes referred to as lenticular galaxies). These galaxies seem to share as many properties with elliptical galaxies as they do with spiral galaxies

At the other extreme, the central bulge is small and the arms are loosely wound. In these Sc galaxies, luminous stars and emission nebulae are very prominent. Our Galaxy and the Andromeda galaxy are both intermediate between the two extremes. Photographs of spiral galaxies, illustrating the different types, are shown in Figure 3, along with elliptical galaxies for comparison.

Figure 3: Hubble Classification of Galaxies. This figure shows Edwin Hubble’s original classification of galaxies. Elliptical galaxies are on the left. On the right, you can see the basic spiral shapes illustrated, alongside images of actual barred and unbarred spirals. (credit: modification of work by NASA, ESA)

The luminous parts of spiral galaxies appear to range in diameter from about 20,000 to more than 100,000 light-years. Recent studies have found that there is probably a large amount of galactic material that extends well beyond the apparent edge of galaxies. This material appears to be thin, cold gas that is difficult to detect in most observations.

From the observational data available, the masses of the visible portions of spiral galaxies are estimated to range from 1 billion to 1 trillion Suns (10 9 to 10 12 MSun). The total luminosities of most spirals fall in the range of 100 million to 100 billion times the luminosity of our Sun (10 8 to 10 11 LSun). Our Galaxy and M31 are relatively large and massive, as spirals go. There is also considerable dark matter in and around the galaxies, just as there is in the Milky Way we deduce its presence from how fast stars in the outer parts of the Galaxy are moving in their orbits.

## How Galaxies are Classified by Type (Infographic)

Astronomer Edwin Hubble, after whom the space telescope is named, classified galaxies according to shape.

The Hubble scale chart takes a wishbone, or tuning fork shape. Armless, elliptical galaxies are on the left. Spirals are divided into those with a central bar and those without one. Looser arm windings are toward the right.

Even distant galaxies, seen as they were billions of years ago, fall into the Hubble shape classifications.

The deeper astronomers look into the universe, the more they see that the expansion of the universe has stretched light, shifting it toward the red end of the spectrum. By measuring the amount of redshift, astronomers can determine how far away a given galaxy is.

A map of 220,000 galaxies produced by the 2dF Galaxy Redshift Survey Team shows the universe has a filamentary structure, seen when it is considered on a large scale.

## National Aeronautics and Space Administration

Part I

Why do people put things into classifications or categories? How does this help us?
_____________________________________________________________________
_____________________________________________________________________
_____________________________________________________________________

What are some things we categorize in our daily lives? Why?
_____________________________________________________________________
_____________________________________________________________________
_____________________________________________________________________

• __________________________________________________
• __________________________________________________
• __________________________________________________
• __________________________________________________
• __________________________________________________

Part II
In your groups, look at the images of actual galaxies and suggest answers to the following questions.

Pretend that a NASA astronomer comes to your school and asks you to name the galaxies pictured in the chart based upon their resemblance to common objects. What would you name them? Write your suggestions underneath each picture.

Without using any prior information, how many different types of galaxies are represented in these pictures? Decide on how many groups or classifications you would have and give each group a name. Then, underneath, include the criteria you would use to include a galaxy in this group.

• Group Name : _______________________________________
Criteria : __________________________________________
_________________________________________________
• Group Name : _______________________________________
Criteria : __________________________________________
_________________________________________________
• Group Name : _______________________________________
Criteria : __________________________________________
_________________________________________________
• Group Name : _______________________________________
Criteria : __________________________________________
_________________________________________________
• Group Name : _______________________________________
Criteria : __________________________________________
_________________________________________________

Now imagine that the NASA astronomer needs your help to classify these newly discovered galaxies based upon your knowledge of the Hubble Fork Diagram. Classify each galaxy according to that scheme. Write the galaxy type and classification below your name of each image. (For example, the Andromeda Galaxy is Spiral, Sb.)

In your groups, research one of the different types of galaxies in the Hubble Fork Diagram. Using the resources provided by your teacher, identify the following information about your galaxy type and present this information to the class.

• Type and classification
• Shape
• Examples of this galaxy type
• How this galaxy forms
• How stars move in this galaxy type

You must use at least one of the following methods to present this information: create one large 3-D poster, write and perform a skit, write and perform a song or rap, or create a 3-D model.

## Classification in Astronomy: Galaxies vs Quasars

Machine Learning and Astronomy go together beautifully. Several astronomical problems involve solving large classification problems with a minimal amount of information. Today, we are going to explore an interesting problem: how to tell if an object is a galaxy or a quasar (an active galactic nucleus accreting an extreme amount of material). Although these two objects are clearly distinct, they often show up as unresolved point sources on telescopes! Let’s explore how we can distinguish between the two using colors! Colors are the difference between two photometric bands (usually in units of magnitude). We are going to be using colors from the Sloan Digital Sky Survey (https://www.sdss.org/). What are the bandpasses involved?

We are going to be using all five bandpasses noted above in the following color vector [u-g, g-r, r-i, i-z]. Finally, we will be using a wonderful set of data from AstroML (https://www.astroml.org/)

from matplotlib import pyplot as plt
from astroML.datasets import fetch_sdss_galaxy_colors
import numpy as np
from keras.models import Sequential
from keras.layers import Dense, InputLayer, Flatten, Dropout
from sklearn import preprocessing
from sklearn.metrics import confusion_matrix
import seaborn as sns
import random

Now that we have the imports, let go ahead and download the data and set up our color vector.

data = fetch_sdss_galaxy_colors()
data = data[::10] # truncate for plotting

# Extract colors and spectral class: [u-g, g-r, r-i, i-z]
ug = data[‘u’] — data[‘g’]
gr = data[‘g’] — data[‘r’]
ri = data[‘r’] — data[‘i’]
iz = data[‘i’] — data[‘z’]
color_vec = np.array([ug, gr]).T#, ri, iz]).T
spec_class = data[‘specClass’]

The spec_class vector contains the classification labels. We have two options: Galaxy or QSO.

We can take look at the colors (or at least a simple 2D projection).

# Let’s take a quick look at some of the colors!
with plt.xkcd():
fig = plt.figure()

ax.set_xlim(-0.5, 2.5)
ax.set_ylim(-0.5, 1.5)

ax.scatter(ug[galaxies], gr[galaxies], c=’b’, label=’galaxies’)
ax.scatter(ug[qsos], gr[qsos], c=’r’, label=’qsos’)

ax.legend(loc=2)

ax.set_xlabel(‘$u-g$’)
ax.set_ylabel(‘$g-r$’)

plt.show()

As we can see, galaxies and quasars naturally segregate themselves in color space :) This should make our classification algorithm have an easier job! Let’s see if it does!

We will first set our training and test sets (including labels). Then we need to transform our labels into numbers. I have written out each step here, but normally we should use sklearn. I repeat I’ve done this here solely for pedagogical reasons :)

train_per = int(0.8*len(color_vec))
#Let’s make our training and test sets
X_train = color_vec[:train_per]
Y_train = spec_class[:train_per]
X_test = color_vec[train_per:]
Y_test = spec_class[train_per:]

Y_train_enc = []
for val in Y_train:
if val == ‘GALAXY’:
Y_train_enc.append(0)
else:
Y_train_enc.append(1)
Y_test_enc = []
for val in Y_test:
if val == ‘GALAXY’:
Y_test_enc.append(0)
else:
Y_test_enc.append(1)

We are now going to build a basic neural network with two hidden layers containing 100 nodes each with a dropbox layer in between

activation = ‘relu’ # activation function
initializer = ‘normal’ # model initializer
lr = 0.001 # initial learning rate
loss_function = ‘sparse_categorical_crossentropy’
metrics_ = [‘accuracy’, ‘mae’, ‘mse’]

model = Sequential([
Dense(units=100, kernel_initializer=initializer, activation=activation),
Dropout(0.2),
Dense(units=100, kernel_initializer=initializer, activation=activation),
Dense(2, activation=’softmax’),
])

model.compile(optimizer=optimizer, loss=loss_function, metrics=metrics_)

Excellent! We can go ahead and fit our model :)

With our fitted model, we can immediately calculate our predictions

And let’s see how it looks :)

predictions = [np.argmax(pred) for pred in test_predictions]

# Normalize confusion matrix
matrix_conf = matrix_conf.astype(‘float64’)
norm_fac = np.sum(matrix_conf[:], axis=1)
for row in range(matrix_conf.shape[0]):
matrix_conf[row,:] = np.round(matrix_conf[row,:]*(1/norm_fac[row]),3)*100
# Plot confusion matrix
sns_plot = sns.heatmap(matrix_conf, annot=True, cmap=’GnBu’, xticklabels=[‘GALAXY’, ‘QSO’], yticklabels=[‘GALAXY’, ‘QSO’])
#sns_plot.set(xticklabels=np.arange(1,5))
#sns_plot.set(yticklabels=np.arange(1,5))
plt.ylabel(‘True’, fontsize=’x-large’)
plt.xlabel(‘Predicted’, fontsize=’x-large’)
sns_fig = sns_plot.get_figure()

We’ve seen how we can easily apply machine learning techniques to solve astronomical classification problems. The classification can be improved by playing with the network.

Hopefully, this has been instructional and will encourage you to find more machine learning and astronomy crossovers!

## What are the criteria for the classification of galaxies? - Astronomy

The Hubble classification of galaxies, also referred to as the ‘tuning fork’ diagram because of its shape, classes galaxies along three main lines into:

Edwin Hubble originally identified an evolutionary sequence for the galaxies (from early-type to late-type) as one moved from left to right across the diagram. Although this is now known to be a false interpretation, the terms ‘early-type’ and ‘late-type’ are still used regularly by astronomers in the manner described below, and when discussing broad galaxy types.

Hubble’s elliptical galaxies were classed according to the ellipticity of the galaxy, and given a Hubble type:

where a = semi-major axis and b = semi-minor axis of the ellipse. Observed values range from E0 (circular cross section – a spherical galaxy) to E7 (the most flattened). E0 are considered ‘early-type’ ellipticals and E7 are ‘late-type’ ellipticals.

Located in the fork of the Hubble classification diagram and intermediate between the elliptical and spiral galaxies are the S0/SB0 galaxies. These galaxies show prominent bulges, but no spiral arms.

Spiral galaxies are classed from ‘early-type’ to ‘late-type’ according to the ratio of the luminosity of the bulge compared with the disk, and the amount of winding of the spiral arms. Type Sa (early) spiral galaxies have prominent bulges (bulge to disk luminosity

0.3), tightly wound arms (pitch angle

0.6), and the stars in the spiral arms are distributed very smoothly. Type Sc (late) spirals have the least prominent bulges (bulge-to-disk luminosity

0.05), and loosely wound arms (pitch angle

0.18) that are resolved into clumps of stars and HII regions.

In bars, the spiral arms do not start directly from the bulge, but from an extended bar of stars that passes through the bulge. They share the same range of classifications as non-barred spirals – from Type SBa (early) to SBc (late) – according to the prominence of the bulge and the winding of the spiral arms.

A later extension to the Hubble classification was the inclusion of irregular galaxies in two classes:
Irr I included irregular galaxies that showed some hint of organised structure (such as the Large and Small Magellanic Clouds), while Irr II were those irregulars that were completely disorganised.

Study Astronomy Online at Swinburne University
All material is © Swinburne University of Technology except where indicated.

## Galaxy Classification

Most of the effort toward galaxy classification in this century has consisted of studying photographic images of galaxies by eye. The dominant paradigm is one established by Edwin Hubble (1936), and later refined by Alan Sandage (1961). This classification is referred to as the Hubble Sequence. There have been many subsequent variants, but largely the classification has been based on subjective, qualitative assessments of the distribution of optical light -- what is often referred to as 'galaxy morphology.' There are many problems with the approach, yet it has remained paramount for over 70 years. Some astronomers have realized that categorizing galaxies by their spectra, as was done for stars, might provide an alternative, and possibly more quantitative and physically insightful method of classification. The most notable, and pioneering effort was by W. W. Morgan, at Yerkes Observatory (Morgan & Mayall 1957). Not coincidentally, it was Morgan who quantified and developed the stellar spectral classification scheme you used in Lab #3 (Spectral Classification of Stars).

Ultimately, the value of a classification scheme is its utility. To understand this, one must ask for what purpose, or application is the classification being done? As we will see, there are times when one might want to classify by morphology, and others when using spectra are the only sensible way to proceed.

In this lab you will become expert galaxy classifiers by both methods -- morphology and spectra. You will devise your own classification schemes based on nearby galaxies, and test them against real data. Finally, you will assess which of these two methods might be the most useful in classifying galaxies at the farthest reaches of the universe.

### Section 1 - Creating your own classification scheme

Your first task it to take these 9 images and sort them, by eye, according to some criteria which will become your classification scheme. The goal here is to have a sequence of objects that will form a reference. Using this sequence, in Section 3 you will attempt to classify other galaxies by asking where they best fit in the sequence. Think ahead a little bit here. Since no two galaxies are identical, when you classify other galaxies by comparing them to your reference sequence, you will have to estimate whether they are closer to one 'type' or another. Often they will fall in-between. There may be times when you think a galaxy is extreme, i.e. lies beyond, or outside the range of your reference sequence. Your reference sequence should be designed to allow you to do make these assessments.

Your reference sequence should be one-dimensional, that is, a linear sequence where you line them up in a row. In principle, you might come up with a two-or-more dimensional grid, where you would have several sequences. However, for the current exercise, design a one-dimensional sequence to simplify comparison with results below. Note that your sequence need not have 9 separate categories if you do not think you can tell the difference between some of the galaxies. In other words, you can lump galaxies together into one group, or 'type,' if you feel that distinguishing between them is not warranted.

Try to forget what you have been taught about the Hubble classification scheme. Be creative and realize there is no 'truth,' only utility. Here, the utility here will be how well you can classify other galaxies.

Load the Netscape browser by clicking on the icon. Select the bookmark for this galaxy classification lab, or open a new location for the following url:
http://www.astro.wisc.edu/

mab/education/astro113/galclass_lab.html You should now have this lab displayed on your browser.

View the montage of images as displayed on this page or click on [B]=big, [S]=small, or [N]=negative images in pop-up windows. You can arrange these pop-up windows with your cursor to form your classification sequence. On the MACs in 5517 you may find the pop-up windows cumbersome they will take time to load and the screens are somewhat small. You can resize the pop-up windows by clicking on their lower-right corner and dragging the corner to the desired size.

 NGC 2775 [B] [S] [N] NGC 2903 [B] [S] [N] NGC 3077 [B] [S] [N] NGC 3147 [B] [S] [N] NGC 3368 [B] [S] [N] NGC 4406 [B] [S] [N] NGC 4449 [B] [S] [N] NGC 4559 [B] [S] [N] NGC 5248 [B] [S] [N]

Credits: CCD Images (Bj or g band) from Frei et al. (1996)
Q1: In your lab book, draw sketches of your reference sequence. This should be a diagram that defines your classification. Be sure to indicate which reference galaxies belong to what stage in your sequence. The 'NGC' labels above each image are catalogue names that can be used for identification.

Devise a labeling scheme for reference and for later classification. When choosing the labeling scheme, keep in mind that you will have to define a classification for other galaxies that might not exactly be in one category or another.

Summarize the criteria you have chosen to order your reference sequence. What are the trend(s) along the sequence? The more careful and specific you can be here, the better you will be able to classify galaxies in the next section. Do you think the trends are physical -- do the represent true physical differences between the galaxies? Or are some of the trends apparent -- do they represent differences in the way you have observed physically similar galaxies? For example, did you choose elongation as a classification parameters? If not, comment on why you did not choose this attribute.

Remove the pop-up windows from your screen. On the MACs in 5517 running JPEGView, click in the box in the upper-left corner. On unix machines running xv, type 'q' in each pop-up window.

### Section 2 - Spectra vs. morphology: which is better?

The spectra are plotted as log(flux) versus wavelength (in Angstroms). The log scale on the y-axis allows you to see the often large changes in flux from wavelength to wavelength. The flux scale (y-axis) has been arbitrarily set to 1 (log(flux) = 0) at 5500 Angstroms.

As you are classifying these spectra, you won't be able to identify the elements responsible for the emission and absorption lines. For reference, then, note the approximate wavelengths of the key features you are using to classify. No doubt you will want to note if the features are in emission or absorption. Again, keep in mind that your sequence need not have 9 separate categories if you do not think you can tell the difference between some of the galaxies' on the basis of their spectra.

To see the individual spectra in more detail, click on them. Remove the pop-up windows as before.

 NGC 2775 NGC 2903 NGC 3077 NGC 3147 NGC 3368 NGC 4406 NGC 4449 NGC 4559 NGC 5248

Credits: Spectra from Kennicutt (1992).
Q2: Construct a spectral classification sequence, noting key spectral features for each point in your classification sequence.

Comment on whether you think there is more or less likelihood of this classification representing true physical differences in the galaxies. For example, if you viewed a spiral galaxy from different angles, what do you think would change more: it's visual appearance (morphology), or its spectrum?

Recall that in Lab 3 the spectrum of a star allowed you to identify what kind of star it was. Knowing this, what can you infer about the content of a galaxy from its spectrum? Can you make as direct an inference about the content of a galaxy from its morphology?

Comment on how well the two methods compare. Does one scheme seem better able to distinguish between this set of galaxies? Do you think one scheme might be more accurate or quantifiable than the other?

Remove any pop-up windows from your screen.

### Section 3 - Testing your schemes against other galaxies

 NGC 3623 [B] [S] [N]

Credits: CCD Images (Bj or g band) from Frei et al. (1996) spectra from Kennicutt (1992).
Q5: When you are done, ask your instructor to identify which galaxies in your reference sequence have the most similar Hubble type.

How well did you match up NGC 3623 to galaxies of comparable Hubble type for your two classification schemes? Did one scheme work better than the other? Explain what you expected.

Next, try four other galaxies, some of which are more distant than those you examined in Section 1. The images here are based on photographic plates which, in this case, have coarser spatial sampling than the CCD images. These two effects (greater distance and poorer sampling) combine to make the images less well resolved, i.e. more blurred and coarser. Another effect of the photographic plates is that they have less dynamic range than the CCD images in Section 1. In other words, it is not possible for the photographs to record the faint, outer parts of the galaxies as well as the bright cores, or vice versa. In these images the cores appear to be 'burned out,' or saturated.

 A B C D

Credits: Digitized photographic images from POSS-I
Q7: Classify these same 4 'mystery' galaxies via their spectra alone. The galaxies are intentionally given different labels here, 1-4. Are the spectra of comparable quality as the ones you used in Section 2 for your spectral classification?

 1 2 3 4

Credits: Spectra from Kennicutt (1992).
Q8: Make your best estimate of matching A-D with 1-4. Estimate (very roughly) on the reliability of this matching. You might base this, for example, on how closely the reference galaxies lined up when ordered in both classification sequences.

Show your results to your instructor and have him/her give you the correct matches. Did you do as well as you expected?

Does it make more sense to use one classification scheme here over the other?

## Galaxies

Like stars and nebula, galaxies emit radiation that can be seen in the electromagnetic spectrum. This includes visible light, radio waves, ultra-violet, infrared, x-rays and gamma rays.

An Active Galactic Nucleus (AGN) is an area at the centre of a galaxy that has an above average brightness (luminosity) over the spectrum. These galaxies are called active galaxies.

The area at the centre is powered by a supermassive black hole. The amount of material spinning around the black hole forms an accretion disc. The heat caused by its speed and the effect of it falling into the black hole produces enormous radiation across the electromagnetic spectrum.

Scientists cannot see the black hole but they can sometimes see radiation at different wavelengths forming an accretion disc, or by looking for jets of material thrown out of the galaxy by the gravity of the black hole.

There are 3 types of active galaxy we shall look at:

##### Seyfert

These have bright spectral emission lines caused by either the accretion disc, or from highly ionized gas around the nucleus. Gas that rotates the black hole faster shows a broader emission line. Typically, serfert galaxies are spiral or irregular.

##### Blazers

Blazers are active galaxies that have jets pointing towards Earth. This is the effect of 'looking down' the jet of a blazer. It is difficult to make comparisons between blazers because of the angle from which we view them. When they are angled at 90 to 35° we see them differently to when they're angled between 0 and 35° to our line of sight. Because of this they are quite variable, emitting radio waves and x-rays.

##### Quasars

Quasars are active galaxies emitting radio and x-rays. They have extremely large black holes at their centre and are moving away from us at extremely fast rates. They are the furthest and oldest objects we know of some are as far as 11 billion light years away.

## Other Ways to Classify Galaxies

When you look at millions of galaxies as the SDSS does, you can't classify every one by looking at it and placing it on the Hubble Tuning Fork. If it takes 30 seconds to find and classify a galaxy on the Hubble Tuning Fork, it would take almost 100 years to classify all the galaxies in the SDSS's Data Release 5! To classify all the galaxies, astronomers need a faster method.

Fortunately, you can use other properties of galaxies to classify them. Astronomers have known for a long time that galaxy type and color are related. Spiral galaxies tend to have more star forming regions, and younger, bluer stars. Elliptical galaxies tend to have mostly old, red stars.

A team of SDSS astronomers looked at the colors, spectra, and visual images of many galaxies to determine how colors are related to galaxy types in SDSS data. The researchers found that galaxies fell into the clearest groups when they looked at the difference between the ultraviolet (u) and red (r) filters. Specifically, the researchers found that most early galaxies (elliptical, SO, and Sa or SBa) had a u-r value greater than 2.22, and that most late galaxies (Sb or SBb, Sc or SBc and Irregular) had a u-r value less than 2.22.

The method of classifying galaxies by their colors is not perfect. There are some unusually red spiral galaxies and some unusually blue elliptical galaxies. However, the method works well enough that it can be used to analyze the properties of large numbers of galaxies fairly easily.

Exercise 4. Look up the following galaxies in the Object Explorer by clicking on their object IDs in the table below. Classify them as early (E, S0, Sa or SBa) or late (Sb or SBa, Sc or SBc, Irr) galaxies from their u-r values (u and r are located to the right of the galaxy's image). Then, look at their images and classify them on the Hubble tuning fork. (Note: you may need to click on the image and open the Navigate tool to get a better view of the galaxy.)