Astro 501, Fundamental Astronomy Problem Set 7

Due December 9

1. An Sc disk galaxy has the following properties:

I-band major-axis disk scale length . . . 50′′
I-band central surface brightness . . . . . 20.2 mag arcsec−2
Disk axis ratio . . . . . . . . . . . . . . . . . . . . . . . . 0.4
Observed 21-cm line width . . . . . . . . . . . . 400 km s−1
Assume that the I-band Tully-Fisher relation relates the absolute I-band magnitude of a
galaxy to the velocity width of its 21-cm line emission by
MI = −8.72(log W − 2.50) − 20.94
If there is no foreground reddening, what is the distance to the galaxy? (Hint: before you go
ahead with the calculation, think carefully about the best way to integrate the luminosity
distribution of an intrinsically circular, but observationally inclined galaxy.)

The inclination of an intrinsically flat galaxy is given by cos i = b/a, so i = 66.4◦ . The true
line-width of the galaxy is then
W = Wobs / sin i = 436 km s−1
Plugging this into the Tully-Fisher relation yields an absolute magnitude of MI = −22.16.

Next, you need to integrate to get the galaxy’s apparent luminosity. If the galaxy were
face-on, this integral would simply be
∫ ∞
Ltot = 2πrI(r)dr
0
However, since the galaxy is inclined, its observed central surface brightness is more than its
true central surface brightness by a factor of cos i. (Due to inclination, the same amount of
light is being squeezed into a smaller area.) So the total luminosity of the galaxy is
∫ ∞ ∞

−r/r −r/r 
Ltot = 2πrI(0) cos i · e d dr = −2πI(0) cos i · (r + rd )e d  = 2πI(0)rd2 cos i
0 
0

So the apparent magnitude of the galaxy is

( )
mI = m(0) − 2.5 log 2πrd2 cos i = 20.20 − 9.50 = 10.70
The distance modulus is then
(m − M )0 = mI − MI = 32.86 = 5 log d − 5 =⇒ d = 37.4 Mpc
2. Consider a spherically symmetric virialized galaxy cluster that is 16 Mpc away. This
cluster contains X-ray emitting gas which we can assume is in hydrostatic equilibrium. The
galaxies in the cluster have a line-of-sight velocity dispersion, σ = 900 km s−1 .

a) In the absense of other heating or cooling mechanisms, what would be the temperature of
the gas which is lost from the galaxies and is thermalized in the intracluster medium? (For
simplicity, just assume the mean molecular weight is one.)

If the galaxies’ velocities are turned into thermal motion, then

1 3
µmH ⟨v 2 ⟩ = kTgal
2 2
where µ is the mean molecular weight of the gas (a number that is close to 1). Since we
only measure one component of the velocity dispersion

1 3 µmH σ 2
µmH 3σ 2 = kTgas =⇒ Tgal = = 1 × 108 K
2 2 k

b) The X-ray gas doesn’t have to be at the same temperature of the galaxies: infalling
gas and supernovae may heat the gas, while radiative processes will cool it. Suppose the
temperature of the gas is measured to be ∼ 3 × 107 K, and imaging of the gas shows that
the x-ray emission declines with radius as ϵx ∝ r−1.6 . If the gas is roughly isothermal, what
is the total mass contained within the central 1 degree of the cluster?

The mass of an x-ray cluster is given by

{ }
kT d ln ρ d ln T
M(r) = − + r
µmH G d ln r d ln r

Since the gas is roughly isothermal, the temperature derivative is zero, so

kT d ln ρ
M(r) = − r
µmH G d ln r

Also, since X-ray emission goes as the square of the density,

d ln ρ
ϵx ∝ ρ2 ∝ r−1.6 , =⇒ ρ ∝ r−0.8 =⇒ = −0.8
d ln r
At 16 Mpc, 1◦ corresponds to r = 16 Mpc × sin(1◦ ) = 0.28 Mpc. So, plugging in the
numbers then gives M ∼ 2.6 × 1046 gm or 1.3 × 1013 M⊙ .
3. These days, finding extrasolar planets is very popular. So let’s look for them in the halo
of another galaxy! The spiral galaxy M31 is at a distance of 770 kpc and has an inclination
of 77◦ . Like most spirals, M31 has a flat rotation curve (vrot = 250 km s−1 ), which suggests
a spherically symmetric dark matter halo with an 1/r2 density law.) Let’s assume this halo
extends out to a distance of Rmax = 30 kpc.

a) Assume that the dark halo of M31 is primarily composed of MACHOs (MAssive Compact
Halo Objects), which could be a collection of black holes, dead neutron stars, brown dwarfs,
or even planets. (Big Bang nucleosynthesis models suggest that this isn’t the case, but
let’s try it out anyway.) For microlensing to work (or, at least obey the amplification
equations), a source must fit entirely within the Einstein ring of the lensing object. So, what
is the minimum mass that a distant M31 halo object must have to microlens a typical 1 M⊙
main-sequence star in M31’s disk? Could you detect an Earth-like planet? Justify your
assumptions for this order-of-magnitude calculation.

The size of the Einstein ring is given by

( )1/2
4GM Dds
α0 =
c 2 Dd Ds

If the entire star is to be lensed, α0 > R/Ds , where R is the stellar radius. For objects in
M31’s halo, Dd ≈ Ds = 770 kpc. Now if we adopt Dds ≈ 30 kpc as a typical distance
between the source and the lens, then

α02 c2 Dd Ds R 2 c2
M= ≈
4GDds 4GDds

which, when you plug in the numbers gives

( )2
R
M = 8.86 × 10−8 M⊙
R⊙

For a solar-type star with R = 1 R⊙ , this is about 0.03 Earth masses, or 2.4 lunar masses.
b) How long will a typical microlensing event last if the size of the Einstein radius is greater
than the size of the background star.

The length of the lensing event for an object passing through the center of an Einstein ring
is given by
( )1/2
R vt1/2 4GM Dds
α0 = = =
Ds Ds c 2 Dd Ds
Plugging in the numbers gives t1/2 = 9.3 × 106 sec, or 108 days. Since R is a radius, the
total time is twice this, or 216 days.

(Note for extra credit: this is the maximum time.

s
To find the duration of a “typical” lens event,
b consider that the √ chord an object will traverse
R has a length s = R2 − b2 , where b is the im-
pact parameter. Also, note that the frequency
of lensing events, as a function of impact param-
eter will go as 2πbdb. Thus the typical event will
last
]R
∫R 2
(R − b 2 )1/2 · 2πbdb 1/3(b2 − R2 )3/2
0 = 0 = 2 R or two thirds as long).
∫R ]R
0 2πbdb 1/2b2 3
0
c) Now consider the probability of actually observing a microlensing event caused by an
M31 MACHO. Let’s define M31’s “optical depth” to lensing as the probability of an object
falling within the Einstein ring of some halo lens positioned somewhere along the line of
sight. Derive this value as a function of the mass of the lens and position along the minor
axis of the galaxy. Are you more likely to see micro-lensing events from high-mass MACHOs
or low-mass MACHOs? (Note that, since the probability of any specific lensing event is
small, the total probability of observing any lensing event is simply proportional to the total
area covered by all MACHOs along the line-of-sight.)

x1
x2

r
b
θ R max
90−ι

M31 disk

M31 Halo

The area lensed by any one MACHO is σ = πα02 Ds2 , or, letting Dd ≈ Ds , and ℓ = Dds
{ }
4Gm ℓ 4πGm
σ = πDs2 = ℓ
c2 Ds2 c2

where m is the mass of the lensing MACHO. If the density of particles is n(r) = n0 /r2 then
the total halo mass (out to a distance Rmax ) is
∫ R ( )
max n0
Mtot = 4πr2 m dr = 4πn0 m Rmax
0 r2
Assuming that the total area of the galaxy that is lensed at any one time is small (i.e., that
the lenses don’t overlap), the total area lensed along a given line-of-sight is
∫ x ∫ x
2 2 Mtot 1 4πGm GMtot ∫ x2 ℓ
τ= n(r)σdℓ = ℓ dℓ = dℓ
x1 x1 4πmRmax r2 c2 Rrax c2 x1 r2
Note that the result is independent of lens mass!
Let’s now define θ as the angle between the radius vector to a point, and the normal to the
line-of-sight. From the figure, we have
r = b sec θ
ℓ = b tan i + b tan θ
dℓ = b sec2 θdθ

where i is the inclination of M31’s disk, and b is the observed distance along the minor axis.
Note that θ can be either positive or negative. The optical depth is now
GMtot ∫ x2 ℓ
τ= 2 dℓ
c Rmax x1 r2

GMtot ∫ arccos(b/Rmax ) b sec2 θ

= 2 b(tan i + tan θ) dθ
c Rmax −i (b sec θ)2
{ }
GMtot ∫ arccos(b/Rmax ∫ arccos(b/R
max )
= 2 dθ + tan θdθ
c Rmax ) −i −i

If we let θm = arccos(b/Rmax ), then

{ ( )}
GMtot Rmax cos i
τ= 2 (θm + i) + log
c Rmax ) b

(One more item to note: in general, the mass and size of any system is related to its rotational
velocity by
v2r
M=α
G
where α is of the order of unity. If you plug this in, then
{ ( )}
vrot Rmax cos i
τ = 2 (θm + i) + log
c b
Written this way, the total mass of the halo drops out, and Rmax only enters in the log.)