The Carina Project.

We have performed a new abundance analysis of Carina red giant (RG) stars from spectroscopic data collected with UVES (high spectral resolution) and FLAMES/GIRAFFE (high and medium resolution) at ESO/VLT. The former sample includes 44 RGs, while the latter consists of 65 (high-resolution) and 800 (medium-resolution) RGs, covering a significant fraction of the galaxy's RG branch, and red clump stars. To improve the abundance analysis at the faint magnitude limit, the FLAMES/GIRAFFE data were divided into ten surface gravity and e ective temperature bins. The spectra of the stars belonging to the same gravity and temperature bin were stacked. This approach allowed us to increase the signal-to-noise ratio in the faint magnitude limit (V ⩾ 20.5 mag) by at least a factor of five. We took advantage of the new photometry index cU,B,J introduced recently as an age and probably a metallicity indicator to split stars along the red giant branch. These two stellar populations display distinct [Fe/H] and [Mg/H] distributions: their mean iron abundances are -2:15 ± 0:06 dex (σ = 0:28), and 1:75 ± 0:03 dex (σ = 0:21), respectively. The two iron distributions di er at the 75% level. This supports preliminary results. Moreover, we found that the old and intermediate-age stellar populations have mean [Mg/H] abundances of -1:91 ± 0:05 dex (σ = 0:22) and 1:35±0:03 dex ( σ= 0:22); these di er at the 83% level. Carina's -element abundances agree, within 1σ, with similar abundances for field halo stars and for cluster (Galactic and Magellanic) stars. The same outcome applies to nearby dwarf spheroidals and ultra-faint dwarf galaxies in the iron range covered by Carina stars. Finally, we found evidence of a clear correlation between Na and O abundances, thus suggesting that Carina's chemical enrichment history is quite di erent from that in the globular clusters. [ABSTRACT FROM AUTHOR]