Name: _______________________________ Lab Partner(s):

________________________________
________________________________
________________________________

Determining Ages of Star Clusters Using Color

Magnitude Diagrams
________________________________________
The Determining Cluster Ages Using CMDs pre-lab exercise is a
prerequisite for this exercise. It should be turned in before beginning.

Lab Exercise
Learning Goals

Describe what is meant by a color-magnitude diagram (CMD).

List the information that can be inferred from a CMD.
Itemize the assumptions made when working with a CMD.
Qualitatively state the relationship between the color of a main-sequence star and its mainsequence lifetime.
Explain the fundamental basis for determining the age of a cluster given the clusters turnoff B V.
Determine cluster ages using their color-magnitude diagrams.
Summarize the complete process of how astronomers estimate the ages of star clusters.

Introduction
Importance of determining the ages of star clusters
Clusters are groupings of stars that were all born at roughly the same time. When we speak of the
age of a cluster, we are equivalently asking the ages of the stars.
So why do we care how old clusters are? The age of a cluster can provide clues to the formation
of galaxies and therefore the Universe. For example, stars in globular clusters appear to be the
oldest in our galaxy. We can then infer that globular clusters were some of the first structures to
evolve during the formation of our galaxy.
Effect of cluster distance
If one star in the sky looks brighter than another, can we say that star is more luminous?
When we look up to the sky at night, we see stars of varying brightness; that is, their apparent
magnitudes are different. This difference is caused in part by the varying luminosities of the
stars. But it could also be caused by the distance the star is from Earth. For example, the Sun
appears brighter to us than Vega, but Vega is more than 50 times as luminous. Thus, we usually
cannot use apparent magnitude of a star as an estimate of its luminosity.
Adapted from University of Washington

In a cluster of stars, however, we can use the apparent magnitude of its constituent stars as a
relative measure of their luminosities. Since a cluster is localized to one relatively small volume
of space, all of the stars within a cluster are roughly at the same distance from Earth. Because the
stars are at roughly the same distance, differences in apparent magnitudes are due only to
differences in the stars luminosities.
Clusters are three-dimensional objects, so there will still be some difference in distance between
Earth and the individual stars in a cluster. However, since a cluster is localized to one (relatively)
small volume of space, the differences in the distances to the stars is negligible. The following
illustration, which is totally not to scale, may help.
To stress this point further, the apparent magnitudes of stars within a single cluster can be used to
compare luminosities among those stars: The lower the apparent magnitude number of a star in a
certain cluster, the greater the luminosity compared to all of the other stars. This is due to the fact
that the distance between individual stars in a cluster is much, much less than the distance
the cluster is from Earth. The distance between the front and back edges of the cluster simply
does not significantly affect the apparent magnitudes of the stars of a cluster, making the
differences in the luminosities of the stars the overriding cause of the different apparent
magnitudes. If a star in a cluster appears brighter to us compared to another star in the same
cluster, then the brighter star must have a higher luminosity.

F
igure 1
Finding the turn-off color-index of a star
cluster
We are finally ready to turn to the object of
interest: color-magnitude diagrams, or CMDs
for short. A CMD, such as the one to the right,
is a plot of all of the stars in a cluster, but with
a very specific set of axes. Along the horizontal
axis is the color-index of each star, and along
the vertical axis is the apparent magnitude of
each star. Note that the vertical axis is
reversed: the values for apparent magnitude get
smaller as you go up. Pause a moment to look at
the CMD to make sure you note each of these
features.

Adapted from University of Washington

Figure 2

Every point on a CMD represents exactly one star. We can read off a stars B V value by
extending an imaginary vertical line down from the star to the horizontal axis. We can read off a
stars apparent magnitude by extending an imaginary horizontal line left from the star toward the
vertical axis.
For practice (be sure to check your results with your instructor to make sure you get them right):
A. Using the CMD in Figure 2, estimate the B V values of stars X and Y.
a. Star X: 0.61 and Star Y: 0.42
b. Star X: 19.5 and Star Y: 20.5
c. Star X: 0.42 and Star Y: 0.61
d. Star X: 20.5 and Star Y: 19.5
B. Which star is more luminous?
a. X
b. Y
c. X and Y have approximately the same luminosity
C. Assuming both X and Y are on the main sequence and were born at the same time, which
star will leave the main sequence first?
a. X
b. Y
c. X and Y will leave the main sequence at about the same time.
D. Which of the following explanations are true.
I.
Higher temperature and higher mass indicate higher apparent magnitude (luminosity)
II.
Lower temperature and lower mass indicate higher apparent magnitude (luminosity)
III.
Small B V value indicates shorter main sequence lifetime
IV.
Large B V value indicates shorter main sequence lifetime
a.
b.
c.
d.

I and III
II and III
I and IV
II and IV

Its possible to visually identify the main sequence stars

on a color-magnitude diagram. They lie within the long,
relatively straight patch of stars that extends from the
bottom right toward the top left of the CMD. Make sure
this agrees with the Main sequence on the CMD to the
right (Fig. 3), where weve drawn a heavy dashed line to
represent the lower edge of the main sequence.
Figure 3

Adapted from University of Washington

E. Put a box around the main sequence on the CMD to the right (this is meant to be easy).
Put a diamond around the group of stars that have left the main sequence and thus are
no longer on it.
The stars you put a diamond around have run out of hydrogen fuel in their cores. That is, they
have spent their lifetime on the main sequence and, in astronomer speak, have left it. The
transition region between stars on the main sequence and those that are already off the main
sequence is of importance to us. It is called the turn-off point or turn-off region. (The location of this
point will vary from cluster to cluster, and will allow us to eventually determine the age of clusters.)

F. Circle the small region on the CMD shown in Fig. 3 above where stars are just
beginning to peel away from the
main sequence. The B V value
corresponding to this turn-off point
is called the B V turn-off point.
What is the B V turn-off point for
this cluster? (You should come
somewhere close to the arrow along
the B V axis.)
G. The approximate turn-off point for a
different cluster, NGC 6397 (Fig. 4,
shown to the right) has been circled
for you. Draw a diagonal line that
represents the lower edge of the main
sequence stars. Estimate the B V
value for the turn-off point for this
cluster: B V =
Figure 4
a.
b.
c.
d.

0.5
1.0
1.5
None of the above

You should now have a good idea of what a CMD plots and how to determine the B V turn-off
point. Be sure to let your instructor know if you are still a bit confused about this process. This
method is something that many astronomers do in their careers today.
Procedure
I. Estimating the ages of clusters
Let us re-focus our attention to individual stars for a moment, instead of entire clusters. Thus far,
we have been very qualitative in our descriptions; we know only that main sequence stars with a
lower B V value will spend less time on the main sequence. There is, however, a quantitative
relationship between these two quantities (see Table 1).
Adapted from University of Washington

Table 1
Color-Index
(B-V)
Main
Sequence
Lifetime
(years)

-0.4

-0.2

0.2

0.5

0.7

1.0

1.6

< 106

3 x 107

4 x 108

4 x 109

1x
1010

6x
1010

> 1011

Take a look at the main sequence lifetime for a star of color-index -0.2. The stated lifetime,
3107 years, is the time it will take a star, from when it first starts fusing hydrogen to helium
in its core, to leave the main sequence. Thus, if a star of color-index -0.2 started its fusion this
morning, it will leave the main sequence in 3107 = 30,000,000 = 30 million years. If a star of
color-index -0.2 started its fusion 10 million years ago, it would leave the main sequence in 20
million years. If the star started its fusion 30 million years ago, it would just now be leaving the
main sequence. (Of course, these are just approximate numbers!)
1. An astronomer measures the visual magnitude of a main sequence star called Keid (located in
the southern constellation of Eridanus) through a blue filter to be 5.4 and the corresponding
visual apparent magnitude value to be 4.4. Once its core fusion starts, how long would it take
Keid to leave the main sequence?
a. 4 x 108 years
b. 4 x 109 years
c. 1 x 1010 years
d. 6 x 1010 years
2. The same astronomer measures the color-index of a different star, Altair (located in the
northern constellation of Aquila), to be 0.2. This astronomer knows that Altair is about 3108
= 300,000,000 years old. Approximately how many years from now will the star leave the
main sequence?
a. 100 million or 1 x 108 years
b. 200 million or 2108 years
c. 300 million or 3108 years
d. 400 million or 4108 years
It must be stressed that the listed main sequence lifetimes in Table I are not the current age of
the main sequence star. Rather, they are the life expectancy of the star for remaining on the main
sequence, regardless of how old the star actually is now. If humans were like stars, every person
on Earth would have a lifetime expectancy written on them. This lifetime expectancy would
not change, regardless of how old the person was. Likewise, the lifetime of a main-sequence star
can be estimated from its B V value.
There is one important property of clusters that allows us to determine the age of the cluster: All
of the stars in a given cluster were born at the same time. This is due to the fact that stars in a
cluster were formed from the same giant gas cloud that collapsed and condensed into stars at the
same time. So all of the stars in a cluster are of the same age. If we could determine the age of
just one of the stars in that cluster, we would know how old all of the stars are. Therefore, we
Adapted from University of Washington

could determine the age of a cluster if we just knew the age of one star. That star needs to be at
or close to the turn-off point of the main sequence.
Keeping this in mind, take a look at the stars at the turn-off point for NGC 6397 in Fig. 4; you
estimated the B V turn-off and should have gotten roughly 0.5.
3. According to Table 1 above, how long do stars of B V = 0.5 live on the main sequence?
a. 4 x 108 years
b. 4 x 109 years
c. 1 x 1010 years
d. 6 x 1010 years
4. Are the stars at the turn-off point leaving the main sequence?
a. Yes
b. No
5. Using this information, how old are the stars at the turnoff point?
a. 4 x 108 years
b. 4 x 109 years
c. 1 x 1010 years
d. 6 x 1010 years
6. How old are all of the stars in this cluster?
a. 4 x 108 years
b. 4 x 109 years
c. 1 x 1010 years
d. 6 x 1010 years
7. How old is NGC 6397?
a. 4 x 108 years
b. 4 x 109 years
c. 1 x 1010 years
d. 6 x 1010 years
Thats it. You have just determined the age of NGC 6397 using only one bit of information: the B
V value of the turn-off point.

__________________________________

We also assume that the stars all have the same composition. This leaves only the differences in
masses that leads to the differences in stars as plotted on a CMD.
8. In your own words, list four of the steps that allow us to determine the age of a star cluster.

Adapted from University of Washington

II. Following ONE cluster through time

Stars within a cluster get older, just as everything else in the Universe does. If we are to believe
that the B V turn-off point can tell us the age of a cluster, then the B V of the turn-off of a
given cluster must change through time.
9. You probably determined above that the age of NGC 6397 was around 4 billion years. In 6
billion years, the age of the cluster will be 10 billion = 11010 years old. What will the B V
turn-off value be for NGC 6397 in 6 billion years?
a. 0.2
b. 0.5
c. 0.7
d. 1.0
10. Generalize your result from question 5: Suppose we you were able to witness the entire
evolution of a cluster of birth. Describe how the B V turn-off would change as time
progressed.

11. Examine the four simulated CMDs that are shown on the next page. These CMDs all
represent the same cluster of stars, but at different ages. (We would have to observe this
cluster for billions of years in order to get the data to produce these CMDs. This, obviously,
is impossible as humans live for only ~100 years, the high-tech astronomy we use today is
less than 50 years old, while the age of the Universe and thus the oldest clusters would be
around 13 billion years). Circle the turn-off point and measure the B V turn-off value of
each snapshot in time of the cluster. Rank the plots of the CMDs in terms of age of the
cluster when the data was taken, from when the cluster was the youngest to when it was the
oldest.
a. A B C D
b. D B C A
c. B C A D
d. A B D C
e. C A B D
12. Explain the logic you used.

Adapted from University of Washington

13. Instead of the four CMDs representing the same cluster as observed at different ages, suppose
they are representing different clusters that we observed all last night. That is, these CMDs
were plotted yesterday, and all are of different clusters in our galaxy. Rank the CMDs in
terms of the ages of the different clusters shown, from youngest to oldest:
a. A B C D
b. D B C A
c. B C A D
d. A B D C
e. C A B D
14. Explain the logic you used.

15. Determine the open cluster with the given B V of the turn-off (TO) and the given age.
TO 0.1 Age 2 x 108 years
a. Pleiades
b. M 67
Adapted from University of Washington

c. M 15
d. M 55
16. Determine the globular cluster with the given B V of the turn-off (TO) and the given age.
TO 0.3 Age 1.6 x 109 years
a. Pleiades
b. M 67
c. M 15
d. M 55
17. Determine the globular cluster with the given B V of the turn-off (TO) and the given age.
TO 0.4 Age 2.8 x 109 years
a. Pleiades
b. M 67
c. M 15
d. M 55
18. Determine the open cluster with the given B V of the turn-off (TO) and the given age.
TO 0.5 Age 4 x 109 years
a. Pleiades
b. M 67
c. M 15
d. M 55

Pleiades

Adapted from University of Washington

M 67

M 15

M 55

19. In your own words, explain why a color-magnitude diagram allows us to determine the age
of a cluster. This should not be a step-by-step how-to (as in question 4). Rather, your
response should incorporate ideas such as distance, luminosity, lifetime of stars, how clusters
are formed, and similar concepts.

Critical Thinking
20. In the note on page 5, we bring up the fact that it is the differences in the masses of the stars
that are the primary reason that the main sequence stars in star clusters are at different
apparent magnitudes and colors (B V) on a CMD. For stars on the main sequence, what
role does mass play in the life expectancy?
a. The most massive main sequence stars have the shortest main sequence lifetimes.
b. The least massive main sequence stars have the shortest main sequence lifetimes.
c. There is no apparent relationship between the mass of main sequence stars and the
main sequence lifetime.

Adapted from University of Washington

10