Atmospheric NLTE models for the spectroscopic analysis of blue stars with winds

V. Complete comoving frame transfer, and updated modeling of X-ray emission

¹ LMU München, Universitätssternwarte, Scheinerstr. 1, 81679 München, Germany
e-mail: uh101aw@usm.uni-muenchen.de
² Departamento de Astrofísica, Centro de Astrobiología (CSIC-INTA), Ctra. Torrejón a Ajalvir km 4, 28850 Torreón de Ardoz, Spain
³ KU Leuven, Instituut voor Sterrenkunde, Celestijnenlaan 200D, 3001 Leuven, Belgium
⁴ Argelander Institut für Astronomie der Universität Bonn, Auf dem Hügel 71, 53121 Bonn, Germany

Received: 22 May 2020
Accepted: 3 August 2020

Abstract

Context. Obtaining precise stellar and wind properties and abundance patterns of massive stars is crucial to understanding their nature and interactions with their environments, as well as to constrain their evolutionary paths and end-products.

Aims. To enable higher versatility and precision of the complete ultraviolet (UV) to optical range, we improve our high-performance, unified, NLTE atmosphere and spectrum synthesis code FASTWIND. Moreover, we aim to obtain an advanced description of X-ray emission from wind-embedded shocks, consistent with alternative modeling approaches.

Methods. We include a detailed comoving frame radiative transfer for the essential frequency range, but still apply methods that enable low turnaround times. We compare the results of our updated computations with those from the alternative code CMFGEN, and our previous FASTWIND version, for a representative model grid.

Results. In most cases, our new results agree excellently with those from CMFGEN, both regarding the total radiative acceleration, strategic optical lines, and the UV-range. Moderate differences concern He II λλ4200-4541 and N V λλ4603-4619. The agreement regarding N III λλ4634−4640−4642 has improved, though there are still certain discrepancies, mostly related to line overlap effects in the extreme ultraviolet, depending on abundances and micro-turbulence. In the UV range of our coolest models, we find differences in the predicted depression of the pseudo-continuum, which is most pronounced around Lyα. This depression is larger in CMFGEN, and related to different Fe IV atomic data. The comparison between our new and previous FASTWIND version reveals an almost perfect agreement, except again for N V λλ4603-4619. Using an improved, depth-dependent description for the filling factors of hot, X-ray emitting material, we confirm previous analytic scaling relations with our numerical models.

Conclusions. We warn against uncritically relying on transitions, which are strongly affected by direct or indirect line-overlap effects. The predicted UV-continuum depression for the coolest grid-models needs to be checked, both observationally, and regarding the underlying atomic data. Wind lines from “super-ionized” ions such as O VI can, in principle, be used to constrain the distribution of wind-embedded shocks. The new FASTWIND version v11 is now ready to be used.