Astron. Astrophys. 338, 413-434 (1998)

6. Conclusions

We have studied models for the long-term evolution of the Galactic disk which include viscous radial gas flows, gas infall, and infall-induced gas flows. The model predictions were confronted with a large set of observational constraints. Hybrid infall/viscous flow models were found to reproduce most of the relevant observations, at least roughly. Especially, the models can explain
- the "conspiracy" between the dark halo-dominated outer rotation curve and the disk/bulge-dominated inner rotation curve,
- exponential radial stellar density profiles independent of the initial disk profile,
- radial profiles of gas and SFR, respectively, which are in agreement with observational constraints over a wide R range,
- the existence of RAGs.
- local constraints, including the ADF, AMR, PDMF, WDLF, and - relation, are at least approximated by the models. Owing to the simple description of the SFR, not all details of the observational constraints are well-fit by the models. This holds especially for the narrow peak in the G dwarf abundance distribution found by Rocha-Pinto & Maciel (1996). We find a generally better agreement with the observations in models starting from a steep exponential initial gas density profile instead of the standard assumption of the density profile of an isothermal sphere.

With typically -0.003 dex kpc^-1, the RAGs predicted by viscous models are small compared to observational results. On the other hand, the available observational data on RAGs in the Galactic disk are still beset with large uncertainties. The model gradients may be enhanced when infall-induced radial gas flows are involved. Moreover, the iron RAG is significantly larger if an exponential initial gas distribution with a short scalelength is adopted instead of an isothermal sphere. The model RAGs are nearly constant with time which contradicts the indications for shallower abundance profiles for older disk stars. Besides the values for the RAGs, the time-evolution of the abundance profiles may provide a crucial test for viscous evolution models.

The viscous radial gas drift generates a central stellar mass concentration in excess of the exponential stellar disk profile. In models with and the standard initial disk, the central mass concentration is consistent with its interpretation as a fraction of the Galactic bulge. The alternative model with a steep exponential initial disk generates a central mass component which may be identified with the whole bulge on the condition that is was formed in a very early evolutionary stage. It is tempting to speculate that it is essentially a large fraction of this initial disk what evolves into the bulge.

Viscous models predict high gas densities in the inner few kpc contrary to what is observed for the disk of our Galaxy. However, this discrepancy may be considerably reduced if the viscosity is temporary strongly enhanced in the inner region, e.g. due to a transient central bar.

We have not yet considered the effects of a star formation threshold and the existence of an outer cut-off in the stellar density profile. These items will be the subjects of a subsequent study on an alternative family of viscous models, which take into account self-regulation of the star formation process.

© European Southern Observatory (ESO) 1998

Online publication: September 14, 1998
helpdesk@link.springer.de