The Family of Galaxies

348-352

There is a great diversity of galaxy types. Figure 13.1 shows a Hubble image revealing a myriad of galaxy shapes and sizes. As in all sciences, classification plays an essential role. How are the different shapes of galaxies "related", does the morphology of galaxies fit a pattern, does our own Milky Way galaxy fit this pattern, can a galaxy "change its shape"? These questions will occupy us here.

Figure 13.1 HST image of the cluster of Abell S0740 - a cluster of galaxies more then 450 million light-years away in the direction of the constellation Centaurus. (credit: NASA, STScI)

Finding Patterns

Figures 13.2a,b show two very different galaxies - a classic spiral and an elliptical galaxy.

Figure 13.2a The great "Pinwheel Galaxy" or M101 (credit: NASA, ESA, K. Kuntz (JHU), F. Bresolin (University of Hawaii), J. Trauger (Jet Propulsion Lab), J. Mould (NOAO), Y.-H. Chu (University of Illinois, Urbana), and STScI)	Figure 13.2b The elliptical galaxy NGC 4660 (credit: NASA, STScI)

A quick inspection of Figures 13.1 and 13.2 will reveal three broad galaxy types (which we will refine further in a moment). Table 13.1 summarizes these three broad classifications:

The Hubble Tuning Fork Diagram

A more refined morphology and one commonly used today was developed by Edwin Hubble. Figure 13.3 shows the Hubble Tuning Fork Diagram in which ellipticals and spirals are classified according to either their elongation or concentration of nuclear region. The spirals branch into two arms - one with a symmetric nucleus and one with galaxies having barred nuclei.

Figure 13.3 The Hubble Tuning Fork classification scheme

Example 13.1 Use the Hubble classification scheme to classify the Milky Way as shown on the left below.

	Solution: The first thing to note is the bar-structure in the nucleus. Next, look at the amount of concentration in the nucleus - it is closest to a "b-type". A good classification would be SBb.

Colours of Galaxies and Selection Effects

Table 13.1 states that 60% or the majority of galaxies are elliptical. However, surveys in the visible part of the spectrum suggest that spiral galaxies are more common. How can these two contradictory claims be reconciled?

Consider another factor: elliptical galaxies have little or no gas and dust while spirals have large "reserves" of gas and dust. Finally, ellipticals consist of primarily population II stars while spirals also have young population I stars.

Example 13.2 How do the facts listed above help resolve the contradiction that while elliptical galaxies are more common we tend to "see" more spiral galaxies?

Solution: The solution comes when we consider the idea of colour of a galaxy. Population II stars are older and will consist primarily of red coloured stars. The blended light that we see will give the elliptical galaxies a more reddish colour. Spiral galaxies, on the other hand contain lots of bright, O and B stars which are both very luminous as well as much bluer. There are two contributing factors that will make it easier to see spirals than ellipticals: Spiral arms are very luminous due to the presence of OB associations and, blue light is scattered less than red light. Both of these factors contribute to making spiral galaxies seem to be more common than ellipticals.

We can easily conclude that spirals are more common than they are because of the reasons given above. This is called a selection effect. Selection effects result when a bias occurs in how data is collected or analyzed and understanding selection effects is a critical part of making sound, scientific inferences.

Just How Many Galaxies are There?

We can't really answer this question. A question we can answer, however, is how many galaxies can we see using present telescopes? Figure 13.4 is one of the deepest images ever made of the night sky. This is the Hubble Ultra Deep Field and was created from over 800 images taken in 11.3 days of observation! There are 8 foreground stars in this image. The rest of the objects are distant galaxies! There are over 10,000 galaxies in this image which covers only a tiny piece (about 1 ten-millionth) of the entire sky!

Figure 13.4 The Hubble Ultra Deep Field image completed in 2004. (credit: NASA, ESA, S. Beckwith (STScI) and the HUDF Team)

Example 13.3 Use the data provided by the Hubble Ultra Deep Field image to estimate a lower limit for the number of galaxies. Why is this an under estimate?

Solution: This is a scaling problem. If 1-ten millionth of the sky shows 10 000 galaxies then a simple estimate is that we would see (10 million) X (10 000) = (10⁷) X (10⁴) = 10 ¹¹ = 100 billion galaxies! The reason that this is an under estimate is that we are looking in only one wavelength region. It is reasonable to expect that we might see more galaxies if we looked with other wavelengths. In fact, in future chapters we will explore this possibility.

The most ambitious project to date to determine how many galaxies can be seen is the Great Observatories Origins Deep Survey or GOODS. It is a multi-wavelength survey using the great observatories (Spitzer, Hubble, Chandra and others). It confirms as reasonable the answer we got in Example 13.2.

Practice

Classify the following galaxies and provide brief reasons for your classification. Use the Hubble calssification scheme.
a		(HST image)
b		(CFHT image)
c		(HST image)
d		(HST image)
e		(HST image)
f		(CFHT image)
What is the difference between an E0 and an E5 galaxy? How could you tell the difference between an edge on Sc galaxy and an E7 galaxy? Two galaxies, one an elliptical and one a spiral, are so far way that you are unable to see very much detail in either. How could you determine which was most likely the spiral galaxy? You are asked to evaluate an astronomy exhibit at a local museum. In the galaxy exhibit you see a beautifully rendered image of a galaxy rotating gracefully like a pinwheel. On what grounds would you inform the curator that this part of the exhibit could be misleading?