Mg line formation in late-type stellar atmospheres

II. Calculations in a grid of 1D models^⋆

Theoretical Astrophysics, Department of Physics and Astronomy, Uppsala University, Box 516, 751 20 Uppsala, Sweden
e-mail: yeisson.osorio@physics.uu.se

Received: 13 July 2015
Accepted: 13 October 2015

Abstract

Context. Mg is the α element of choice for Galactic population and chemical evolution studies because it is easily detectable in all late-type stars. Such studies require precise elemental abundances, and thus departures from local thermodynamic equilibrium (LTE) need to be accounted for.

Aims. Our goal is to provide reliable departure coefficients and equivalent widths in non-LTE, and for reference in LTE, for diagnostic lines of Mg studied in late-type stars. These can be used, for example, to correct LTE spectra and abundances.

Methods. Using the model atom built and tested in the preceding paper in this series, we performed non-LTE radiative transfer calculations in a grid of 3945 stellar 1D atmospheric models. We used a sub-grid of 86 models to explore the propagation of errors in the recent atomic collision calculations to the radiative transfer results.

Results. We obtained departure coefficients for all the levels and equivalent widths (in LTE and non-LTE) for all the radiative transitions included in the “final” model atom presented in Paper I. Here we present and describe our results and show some examples of applications of the data. The errors that result from uncertainties in the collisional data are investigated and tabulated. The results for equivalent widths and departure coefficients are made freely available.

Conclusions. Giants tend to have negative abundance corrections while dwarfs have positive, though small, corrections. Error analysis results show that uncertainties related to the atomic collision data are typically on the order of 0.01 dex or less, although for few stellar models in specific lines uncertainties can be as large as 0.03 dex. As these errors are less than or on the same order as typical corrections, we expect that we can use these results to extract Mg abundances from high-quality spectra more reliably than from classical LTE analysis.