SpringerLink
Forum Springer Astron. Astrophys.
Forum Whats New Search Orders


Astron. Astrophys. 360, 76-84 (2000)

Previous Section Next Section Title Page Table of Contents

7. Discussion and conclusions

In this paper we have investigated the coupling between DM and baryons during an assumed monolithic formation of two galaxies: a giant and a dwarf spheroidal galaxy.

Starting from a homogeneous distribution, baryons sink toward the center of the virialized DM halo due to cooling instability. The smallest object has about 10 independent episodes of star formation [FORMULA] long, while the largest has basically one single burst which is 10 times longer. This difference nicely matches the available observations. Giant elliptical galaxies are indeed old objects dominated by a single burst of star formation (Bender et al. 1992), whereas dwarf ellipticals basically exhibit an irregular and intermittent SF history (Mateo 1998).

A crucial and sensitive aspect of our simulations is that both the giant and the dwarf object have been modeled with the same number of particles, which implies an increase by a factor [FORMULA] in the mass resolution passing from the giant to the dwarf.

To check whether the results for the SF histories depend on the mass resolution of our simulations, the giant galaxy has been re-simulated using 2,000, 20,000 and 200,000 particles, whilst the dwarf galaxy has been re-simulated using 2,000 and 20,000 particles.

This way the mass resolution goes from [FORMULA] to [FORMULA] for the giant galaxy, and from [FORMULA] to [FORMULA] for the dwarf galaxy. The results are summarized in Fig. 2 and Fig. 7, where we compare the SF histories of the two galaxy models at increasing mass resolution.

In both cases we show a reasonable convergence of the results.

As for the giant galaxy (see Fig. 2), we show that the SF history is dominated by a single ancient burst of SF, whose trend does not change too much passing from 2,000 to 200,000 particles. Nevertheless, a more careful analysis shows that a difference emerges passing from 2,000 to 20,000 particles, whereas passing from 20,000 to 200,000 particles does not significantly change the global result.

Although the SF peak is almost the same in all the simulations, the low resolution run produces a burst larger than the other two cases. Correspondingly, the amount of gas consumed (and hence the final mass in stars) passes from [FORMULA] in the low resolution run to about [FORMULA] in the medium and high resolution run. This means that at increasing number of particles, the results asymptotically converges.

These conclusions confirm previous analyses on the performances of the SPH method (Steinmetz & Müller 1993, Thacker et al. 1998). By comparing different SPH implementations they find that the minimum number of particles required to calculate local physical variables in dynamically evolving systems is about 10,000, and that the SPH method can give reasonable results also by using a small number of particles.

On the other hand, changing the resolution by a factor of 10 does not alter the SF history of the dwarf galaxy (see Fig. 7), which exhibits several episodes of SF, although the number and position of the peaks does not coincide exactly.

The basic results of this paper can be summarized as follows:

  • the stars which are formed show a final distribution which is greatly different from the DM distribution and can be represented by a Hernquist profile with a length-scale [FORMULA];

  • a particular aspect of the mass distribution is that the regions inside [FORMULA] are baryon dominated, while the DM is the main mass component at outer radii, as observed in real ellipticals;

  • finally it is worth noting that, in both objects, the final DM velocity dispersion is about 1-2 times the stars velocity dispersion, in agreement with Loewenstein & White III (1999) findings.

In forthcoming papers we are going to analyze simulations which adopt different initial conditions for DM, in order to test the effect of the DM initial properties (density profile, velocity dispersion and so forth) on the final baryon distribution.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 2000

Online publication: July 27, 2000
helpdesk.link@springer.de