The photospheric solar oxygen project^⋆

II. Non-concordance of the oxygen abundance derived from two forbidden lines

¹ Zentrum für Astronomie der Universität Heidelberg, Landessternwarte, Königstuhl 12, 69117 Heidelberg, Germany
e-mail: Elisabetta.Caffau@obspm.fr
² GEPI, Observatoire de Paris, CNRS, Université Paris Diderot, Place Jules Janssen, 92190 Meudon, France
³ LESIA, Observatoire de Paris, CNRS, Université Paris Diderot, Place Jules Janssen, 92190 Meudon, France
⁴ Leibniz-Institut für Astrophysik Potsdam (AIP), An der Sternwarte 16, 14482 Potsdam, Germany
⁵ European Southern Observatory, Casilla 19001, Santiago, Chile

Received: 11 March 2013
Accepted: 24 April 2013

Abstract

Context. In the Sun, the two forbidden [O i] lines at 630 and 636 nm were previously found to provide discrepant oxygen abundances.

Aims. We investigate whether this discrepancy is peculiar to the Sun or whether it is also observed in other stars.

Methods. We make use of high-resolution, high signal-to-noise ratio spectra of four dwarf to turn-off stars, five giant stars, and one sub-giant star observed with THEMIS, HARPS, and UVES to investigate the coherence of the two lines.

Results. The two lines provide oxygen abundances that are consistent, within observational errors, in all the giant stars examined by us. On the other hand, for the two dwarf stars for which a measurement was possible, for Procyon, and for the sub-giant star Capella, the 636 nm line provides systematically higher oxygen abundances, as already seen for the Sun.

Conclusions. The only two possible reasons for the discrepancy are a serious error in the oscillator strength of the Ni i line blending the 630 nm line or the presence of an unknown blend in the 636 nm line, which makes the feature stronger. The CN lines blending the 636 nm line cannot be responsible for the discrepancy. The Ca i autoionisation line, on the red wing of which the 636 nm line is formed, is not well modelled by our synthetic spectra. However, a better reproduction of this line would result in even higher abundances from the 636 nm, thus increasing the discrepancy.