Satellite Galaxies in the Local Group
My current interest is in the analysis of the spatial distribution of satellite galaxies and their relation to current cosmological models.
This is an illustration of the "disc-of-satellites"
(DoS). A projection of the positions of the
Milky Way satellite galaxies is shown. The red dots mark the positions of the
11 classical satellite galaxies that were known until 2005. Since then more
dSph satellite galaxies have been found;
these are shown by the green dots. The sticks mark the distance uncertainties.
On the left-hand side the projection is chosen edge-on to the plane fitted
through the 11 classical satellites, on the right-hand side rotated by 90°.
I did simulations with my code libHyperBox of a dwarf Galaxy without dark matter that orbits about the Milky Way. You can see an animation of what an observer would see here; this little movie shows 10 Gyrs of evolution.
You can find results of my research in the following publications:
Refereed Publications on this topic
- Metz, M.; Kroupa, P.; Theis, C.; Hensler, G. & Jerjen, H., 2009, acc apJ, arXiv.0903.0375
"Did the Milky Way dwarf satellites enter the halo as a group?" - Metz, M.; Kroupa, P. & Jerjen, H., 2009, acc MNRAS, arXiv:0901.1658
"Discs of Satellites: the new dwarf spheroidals" - Metz, M.; Kroupa, P. & Libeskind, N. I., 2008, ApJ, 680, 287; arXiv:0802.3899
"The Orbital Poles of Milky Way Satellite Galaxies: A Rotationally Supported Disc-of-Satellites" - Metz, M. & Kroupa, P., 2007, MNRAS, 376, 387; astro-ph/0701289
"Dwarf-spheroidal satellites: are they of tidal origin?" - Metz, M.; Kroupa, P. & Jerjen, H., 2007, MNRAS, 374, 1125; astro-ph/0610933
"The spatial distribution of the Milky Way and Andromeda satellite galaxies"
Other publications
- Majewski, S.R., Bullock, J., Burkert, A., Gibson, B., Grebel, E., Gnedin, O.Y., Guhathekurta, P., Helmi, A., Johnston, K.V., Kroupa, P., Metz, M., Moore, B., Patterson, R.J., Shaya, E., Strigari, L.E., van der Marel, R.
"Galactic Dynamics and Local Dark Matter" in "SIM Lite Astrometric Observatory" (arXiv:0902.2759)
See also ADS for a list of refereed publications.
Python
Me and Reinhold Schaaf started a python course for astronomers at the AIfA.
libHyperBox
libHyperBox: I have developed a high-performance particle-mesh code for galactic dynamics simulations, fully implemented in C++. In the current development stage it outperforms a comparable Fortran code by a factor of ~15, is more flexible than the original code, and will be integrateable in simulation frameworks, that are based on Python, soon. A GPU accelerated version is also available, which gives you another performance boost. Most recently we also got support by the "NVIDIA Professor Partnership Program".
The following figure shows a performance comparison of the different codes. The line labeled F77 indicates the measured wall-clock runtime of a Fortran implementation, sbppf is my C++ implementation using the host CPUs only. The various lines labels with G80, 2xG92, G200, and 4xG92 show the measured runtime performance of my GPU accelerated code, using the graphic cards shown above.
The measurements were performed on a quad-core machine (2x Opteron 270, 2GHz) with 4GB RAM and two fully featured PCIe x16 buses to connect the graphic cards. For the 4xG92 runs, two Nvidia 9800GX2 were installed, ie. 4 GPUs were used simultaniously.
