Chemodynamical modeling of young disk galaxies

Disk galaxies are very complex systems consisting of several
components like the bulge, the disk and the halo. Additionally,
substructures like extended HI disks, thin and thick
stellar disks are observed. Contrary to elliptical galaxies,
disk galaxies contain a significant mass fraction of gas. It is
therefore an important question how gas physics influences
the evolution of a galaxy, especially in its early state.
To address this question, I have presented a threedimensional
model for the chemodynamical evolution of
galaxies, consisting of a hot and a cold gas phase and stars,
embedded in a dark matter halo. The components interact
with each other through several processes like star formation,
evaporation and condensation, radiation cooling and
dissipation, and mass and energy feedback from the stars.
In this model, the dynamical, star formation and chemical
evolution of the galaxy is followed. This enables comparison
to a variety of observable properties, such as metallicity
distribution or kinematics of the stars. Using the method
of Westera et al. (2002) to calculate colors of the models
at different redshifts including absorption, the models can
also be directly compared to high redshift observations. The
template of the numerical model was the three dimensional
version of CoDEx (Samland et al., 1997). I have written
the program in the F90 standard, which allowed for vectorization
and partly parallelization of the code. Performance
comparisons assure that the program runs very efficiently
on different architectures. Samland & Gerhard (2003) note that the poorly known
cloud dissipation rate, the rate at which the cloudy medium
can lose energy, is very important for galaxy evolution. It
was pointed out in the description of the model that the
cloud dissipation rate is not well-determined and may well
vary between galaxies. To investigate its effect on disk evolution,
I set up a sequence of models, where only the cloud
dissipation efficiency was varied. I found that two evolutionary
paths for the disk evolution exist in the context of
the model, which are summarized below in more detail.
The evolutionary path of a galactic disk depends on whether
it is the gas disk or the stellar disk which first becomes unstable.
The stability analysis was done using the Toomre Q
parameter of the cloudy and the stellar phase.
When the cold gas cools efficiently and drives the instability,
the galactic disk fragments and forms a number of
massive clumps. The stellar disk fragments with the gas because
of the strong gravity of the clumps, which begin to
form stars at a high rate because of their large density. The
clumps then spiral to the center of the galaxy in a few dynamical
times and merge there to form a central bulge component
in a strong starburst. This scenario is similar to that
discussed by Noguchi (1999). In this mode of disk evolution,
the bulge forms rapidly and early; the unstable region
of the disk is completely disrupted, and the disk must be rebuilt by subsequent infall. Because of the starburst origin,
many of the bulge stars formed in this way have large
[�/Fe] abundance ratios.
On the other hand, if the kinetic energy of the cold clouds
is dissipated at a lower rate, stars form from the gas in a
more quiescent mode, while keeping the kinetic temperature
high enough and the gas density low enough to prevent
the gas from becoming dynamically unstable. In this case,
an instability only sets in at later times, when the surface
density of the stellar disk has grown sufficiently high. The
system then forms a stellar bar, which channels gas into the
center, evolves, and forms a bulge whose stars are the result
of a more extended star formation history, i.e., have lower
[�/Fe]. This scenario resembles the evolution described
by Combes & Sanders (1981), Pfenniger & Norman (1990)
and Raha et al. (1991). The comparably long formation
time for bars and bulges in this mode and the weakening
of the bar through inward gas flow in the still gasrich
disks might explain the absence of bars at high redshift
(Abraham et al., 1999). This is also consistent with the
young age inferred for the Galactic bar (Cole & Weinberg,
2002). An example of a nearby galaxy in which the fragmentation
process may be taking place, is the gas-rich, blue starburst
galaxy NGC 7673 (Homeier & Gallagher, 1999). At
high redshift, chain and multi-clump morphological structures,
as well as synchronized colors in clumpy objects, can
be well explained by a fragmented disk model. The model
suggests that these galaxies are in their formation process
and are observed during their relatively short fragmentation
phase, with a high SFR, comparable to the model SFR of
up to 220 M� yr