The benchmark halo giant HD 122563: CNO abundances revisited with three-dimensional hydrodynamic model stellar atmospheres

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaperJournal articleResearchpeer-review

  • R. Collet
  • Ã. Nordlund, Centre for Star and Planet Formation, Niels Bohr Institute and Natural History Museum of Denmark, University of Copenhagen, Øster Voldgade 5-7, DK-1350 Copenhagen, Denmark 0000-0002-2219-0541, M. Asplund, Research School of Astronomy and Astrophysics, Australian National University, Cotter Road, Canberra ACT 2611, Australia, W. Hayek, National Institute of Water and Atmospheric Research, 301 Evans Bay Parade, Hataitai, Wellington 6021, New Zealand, R. Trampedach, Space Science Institute, 4750 Walnut street, Suite 205, Boulder, CO 80301, USA 0000-0003-4034-0416
We present an abundance analysis of the low-metallicity benchmark red giant star HD 122563 based on realistic, state-of-the-art, high-resolution, three-dimensional (3D) model stellar atmospheres including non-grey radiative transfer through opacity binning with 4, 12, and 48 bins. The 48-bin 3D simulation reaches temperatures lower by ˜300-500 K than the corresponding 1D model in the upper atmosphere. Small variations in the opacity binning, adopted line opacities, or chemical mixture can cool the photospheric layers by a further ˜100-300 K and alter the effective temperature by ˜100 K. A 3D local thermodynamic equilibrium (LTE) spectroscopic analysis of Fe I and Fe II lines gives discrepant results in terms of derived Fe abundance, which we ascribe to non-LTE effects and systematic errors on the stellar parameters. We also determine C, N, and O abundances by simultaneously fitting CH, OH, NH, and CN molecular bands and lines in the ultraviolet, visible, and infrared. We find a small positive 3D-1D abundance correction for carbon (+0.03 dex) and negative ones for nitrogen (-0.07 dex) and oxygen (-0.34 dex). From the analysis of the [O I] line at 6300.3 Å, we derive a significantly higher oxygen abundance than from molecular lines (+0.46 dex in 3D and +0.15 dex in 1D). We rule out important OH photodissociation effects as possible explanation for the discrepancy and note that lowering the surface gravity would reduce the oxygen abundance difference between molecular and atomic indicators.
Original languageEnglish
JournalMonthly Notices of the Royal Astronomical Society
Volume475
Issue number3
Pages (from-to)3369-3392
ISSN0035-8711
DOIs
Publication statusPublished - 1 Apr 2018

    Research areas

  • convection, hydrodynamics, line: formation, stars: abundances, stars: atmospheres, stars: individual: HD 122563

See relations at Aarhus University Citationformats

ID: 121837831