2020AJ....159..174T

It has been suggested that Fe abundances of K dwarfs derived from Fe I and Fe II lines show considerable discrepancies, and oxygen abundances determined from high-excitation O I 7771-5 triplet lines are appreciably overestimated (the problem becoming more serious toward lower T_eff), which, however, has not yet been widely confirmed. With the aim of clarifying this issue, we spectroscopically determined the atmospheric parameters of 148 G-K dwarfs (Hyades cluster stars and field stars) by assuming the classical Fe I/Fe II ionization equilibrium as usual, and determined their oxygen abundances by applying the non-local thermal equilibrium spectrum fitting analysis to O I 7771-5 lines. It turned out that the resulting parameters did not show any significant inconsistency with those determined by other methods (for example, the mean differences in T_eff and logg from the well-determined solutions of Hyades dwarfs are mostly <=100 K and <=0.1 dex). Likewise, the oxygen abundances of Hyades stars are around [O/H] ∼ +0.2 dex (consistent with the metallicity of this cluster) without exhibiting any systematic T_eff-dependence. Accordingly, we conclude that parameters can be spectroscopically evaluated to a sufficient precision in the conventional manner (based on the Saha-Boltzmann equation for Fe I/Fe II) and oxygen abundances can be reliably determined from the O I 7771-5 triplet for K dwarfs as far as stars of T_eff >= 4500 K are concerned. We suspect that previously reported strongly T_eff-dependent discrepancies may have stemmed mainly from overestimation of weak-line strengths and/or improper T_eff scale.