Interacting Galaxies

 
When we look in the sky, we see many galaxies which come in pairs and are often morphologically very peculiar.

We have also seen that interactions are often associated with starburst activity.

What's happening here?

How long do collisions take? What are the timescales involved?

Let's calculate a "passing time" for two galaxies, defined (arbitrarily) as Tpass = 6R/v. 

If a galaxy has a size R=20 kpc and is moving past another with a velocity v ~ 350 km/s, then the passing time is

Tpass = 6(20,000 pc)/(350 pc/Myr) = 340 Myr. 
A long time -- we can't watch it happen. So how do we study interacting galaxies?

 

But why the heck do galaxies merge?

Galaxies can orbit each other just like comets orbit the sun. Comets don't spiral into the sun, but galaxies can spiral together to merge. What's the difference?

Dynamical Friction

 
Imagine a massive object moving through a background "sea" of low mass objects. As it moves through, it creates a trailing "wake" -- an excess in the density of the low mass objects behind it. 

Why would this act like a frictional force in the motion of the massive object?

What is the massive object?

What is the sea?
Can we be more quantitative? The frictional force has the form
 
  • M: satellite mass
  • rho: mass density of "sea"
  • v: satellite velocity
  • C: depends on situation...
Why does it have this form?

Let's use this to figure out how long it will take a satellite galaxy to merge with a larger one.
 
Assume the large galaxy has a constant rotation curve. This means that the density distribution looks like this:
Plugging that in for rho, we get the expression for the drag force:
Remember that a drag force on an orbiting object is a torque...
...and that torque is the rate of change of angular momentum
Now, the angular momentum (for a circular orbit) is given by
So taking dL/dt and setting it equal to torque we get
Set it up as an integral over
  • r: initial radius to 0
  • t: t=0 to merging time T
and integrate to get 

Example: How long would it take for the Large Magellenic Cloud to fall to the center of our Galaxy?

M=2x1010 Msun
v=220 km/s
R=50 kpc (=50,000 pc)
C=23

T=1,700 Myr = 1.7 Gyr


(A better estimate -- using a more realistic, elongated orbit -- gives a much longer merger time. Why?)

This expression breaks down if the galaxies are comparable in mass/size, but it's still true that in close collisions, galaxies can merge on timescales of a billion years or so.

How would things be different if galaxies didn't have dark matter?

And whats's with these tails?

Think back to gravitational tidal forces: they act to radially stretch anything passing near a massive object. Couple that with the fact that galaxies are only bound by gravity, and that they are rotating, and we can see that tails for from material "spun off" by gravitational forces during collisions.
 


Look at the GalCrash applet


Making the Cartwheel