2009A&A...505..605C

The evolution of C and O abundances in the Milky Way can impose strong constraints on stellar nucleosynthesis and help in understanding the formation and evolution of our Galaxy. The aim of this paper is to review the measured C and O abundances in the disk and bulge of the Galaxy and compare the results to predictions of Galactic chemical evolution models. We adopt two successful chemical evolution models for the bulge and the disk, respectively. They assume the same nucleosynthesis prescriptions but different histories of star formation. The data show a clear distinction between the trend of [C/O] in the thick and thin Galactic disks, while the thick disk and bulge trends are indistinguishable with a large (>0.5dex) increase in the [C/O] ratio in the range from -0.1 to +0.4dex for [O/H]. In our models we consider yields from massive stars with and without the inclusion of metallicity-dependent stellar winds. The observed increase in the [C/O] with metallicity in the bulge and thick disk lies between the predictions utilizing mass-loss rates of Maeder (1992A&A...264..105M) and Meynet & Maeder (2002A&A...390..561M). A model without metallicity-dependent yields completely fails to match the observations. Thus, the relative increase in carbon abundance at high metallicity appears to come from metallicity-dependent stellar winds in massive stars. These results also explain the steep decline of the [O/Fe] ratio with [Fe/H] in the Galactic bulge, while the [Mg/Fe] ratio is enhanced at all [Fe/H]. We conclude that data and models are consistent with a rapid bulge and thick disk formation timescales, and with metallicity-dependent yields for C and O. The observed too high [C/O] ratios at low metallicity in the bulge may stem from an unaccounted source of carbon: very fast rotating metal poor stars, or metal-poor binary systems whose envelopes were stripped by Roche lobe overflow.