Simulation of interacting galaxies with
,,Minor merger'' (= interaction of a small satellite galaxy with a massive
In a galaxy as our Milky Way each individual star moves about according to the
combined gravitational force from all other stars.
From these gravitational forces the courses of the stars can be calculated.
The computational efforts for doing this are tremendous, since for N stars
N equations have to be solved. For this reason a
Galaxy actually consisting of 100 billion stars is represented by only few
10 thousend testparticles. Measures are taken to compensate for the
"granularity" resulting from the reduced number of particles.
In my project I put up a system of testparticles with initial positions and
velocities as found in our Galaxy. At first I tested whether this system is
stable over a sufficient amount of time (i.e., 1.5 Gyrs). This stable model
was now exposed to a minor merger event, i.e., a second much smaller (spherical)
galaxy was placed on the brink of the larger disc galaxy model with an initial
velocity corotating with the host galaxy and with an additional vertical
velocity component. The following
pictures show how the structures of the host galaxy and of the satellite
galaxy are influenced by the interaction.
1) Development of the structure
Get Mpeg-Versions (more frames):
2) Development of the structure of
the satellite galaxy
The distance of the satellite galaxy to the rotation axis
of the disk galaxy is hardly reduced during the interaction, but reduces
quite rapidly in the last two timesteps. While sinking towards the center of
the disk galaxy the satellite galaxy dissolves.
The development with time of the Lagrange-radii for 0.1%, 0.5%,
1%, 5%, 10%, 20%, 50%, 75%, 90% and 95% of the total mass of the satellite
galaxy and the distance to the rotation axis. The squares mark time steps of
150 Millionen years, respectively.
3) Development of the disk structure
Plot of the development with time of the half mass
height for the outer regions and the center
The vertical structure of the disk changes
due to the interaction. While the inner region doesn't change much, the disk
becomes significantly thicker in the outer regions.
4) Development of the gas dynamic
The dynamic of the gas component is simulated by treating some particles as
,,gas particles''. These particles have the property of sticking together, thus
forming one particle out of two (,,sticky particles''). This merging of
particles is a process which needs a collision, but only collisions between
particles moving at small relative velocities end up in such a merger process.
One therefore distinguises between the collision rate and the merger rate, the
number of particle collisions/mergers per time interval.
In my thesis I added the process of disruption of particles which suffered from
many mergers, thus beeing quite massive. These particles disrupt on time scales
depening on their mass (which can increase due to further merges, while
,,waiting'' for their disruption, but the time scale then further decreases).
Physically this process originates from star formation in massive clouds, and
the cloud material gets driven away by stellar radiation and supernova
explosions. I therefore call the rate connected with these disruptions
Development of the starformation rate (solid line),
the merging rate (dotted line) and the collision rate (dashed line).
For comparison the rates are normalized to their starting values.
The collision rate of gas particles remains constant.
The merging rate and the starformation rate, which is strongly correlated with
the merging rate, increase at first to a constant level, but decrease due to
the enhanced volume as a consequence of the interaction and thus the
smaller collision probability.
Implementation of star formation as fragmentation
of large gas clouds prevents the gas component from becomming too clumpy,
which would lead to a decrease of the collision rate. The constant level
shows that star formation keeps the gas component stable for at least
Diploma thesis(PDF):(German only)
Interaction of galaxies with different sizes
induced by the interaction
photographed by Dr. Uwe Schwarzkopf, 8. Apr.1997
DFOSC at the 1.54-m Danish Telescope at La Silla (Chile),
Exposure time: 30 min in the r-Band.
last modified: Friday, 14-Nov-2008 21:39:03 CET