[Paper Review] The NextGen Model Atmosphere grid for $3000\le \Teff \le 10000\K$

P. H. Hauschildt, E. Baron+1 more|ArXiv.org|Jul 28, 1998

Stellar, planetary, and galactic studiesPhysics and Astronomy2 references644 citations

TL;DR

This paper presents a new LTE model atmosphere grid for stars with effective temperatures from 3000 K to 10000 K, calculated using the PHOENIX code with direct opacity sampling and improved molecular opacities. The grid provides consistent spectral modeling for cool stars and serves as a foundation for future NLTE models, showing good agreement with Kurucz 1994 models at higher temperatures but revealing significant differences at lower temperatures due to enhanced molecular physics and equation of state treatments.

ABSTRACT

We present our NextGen Model Atmosphere grid for low mass stars for effective temperatures larger than $3000\K$. These LTE models are calculated with the same basic model assumptions and input physics as the VLMS part of the NextGen grid so that the complete grid can be used, e.g., for consistent stellar evolution calculations and for internally consistent analysis of cool star spectra. This grid is also the starting point for a large grid of detailed NLTE model atmospheres for dwarfs and giants (Hauschildt et al, in preparation). The models were calculated from $3000\K$ to $10000\K$ (in steps of $200\K$) for $3.5 \le \logg \le 5.5$ (in steps of 0.5) and metallicities of $-4.0 \le \mh \le 0.0$. We discuss the results of the model calculations and compare our results to the Kurucz 1994 grid. Some comparisons to standard stars like Vega and the Sun are presented and compared with detailed NLTE calculations.

Motivation & Objective

• To develop a consistent, high-resolution LTE model atmosphere grid for low-mass stars with effective temperatures from 3000 K to 10000 K.
• To ensure compatibility with the existing VLMS and brown dwarf parts of the NextGen grid for use in stellar evolution and population synthesis modeling.
• To establish a foundation for future full NLTE model atmosphere calculations by using these LTE models as starting points.
• To improve spectral modeling accuracy by incorporating advanced molecular equations of state and opacity treatments not fully present in earlier grids like Kurucz 1994.
• To assess the validity of LTE assumptions in hot stars by comparing LTE and NLTE results, particularly for Vega and solar-type stars.

Proposed method

• Calculated models using the PHOENIX stellar atmosphere code (version 9.1) in static, plane-parallel, LTE mode with full line-by-line opacity sampling.
• Employed direct opacity sampling by dynamically selecting relevant lines from master line lists at each wavelength point during iterations, enabling depth-dependent line profiles and high spectral resolution.
• Used the Kurucz (1994) atomic line list and included detailed molecular opacities and equations of state tailored for cool star atmospheres.
• Applied variable-resolution wavelength grids with up to 204,434 points, including adaptive resolution to resolve strong and NLTE lines.
• Computed NLTE models for Vega (Teff = 9500 K) using 47,593 primary NLTE transitions, with 25,449 lines treated with Voigt profiles and 22,144 with Gaussian approximations.
• Validated results against observed spectra of Vega and the Sun, and compared with Kurucz 1994 models and existing NLTE calculations.

Experimental results

Research questions

• RQ1 How do the new LTE model atmospheres compare to the Kurucz 1994 grid in terms of spectral energy distribution and line formation for solar-type stars?
• RQ2 What are the systematic differences between the NextGen and Kurucz 1994 models at effective temperatures below 5000 K, and what causes them?
• RQ3 To what extent do NLTE effects alter the structure and emergent spectra of hot stars like Vega (Teff ≈ 9500 K)?
• RQ4 How do departures from LTE affect temperature structure and line formation in the outer, optically thin layers of hot star atmospheres?
• RQ5 What is the validity range of the LTE assumption in model atmospheres, and when does NLTE become essential for accurate spectral modeling?

Key findings

• The NextGen LTE model grid shows good agreement with the Kurucz 1994 grid for stars with effective temperatures between 5000 K and 7000 K.
• For temperatures below 5000 K, significant differences emerge due to improved molecular equations of state and more accurate molecular opacity treatments in the NextGen models.
• NLTE effects in the Vega model (Teff = 9500 K) cause temperature structure deviations of nearly 1000 K in the outermost, optically thin layers, indicating strong non-LTE behavior.
• Departure coefficients for ground states of key ions in the Vega model show substantial NLTE departures, especially in the upper atmosphere, confirming that LTE is a poor assumption for hot stars.
• The LTE assumption breaks down significantly above ~7000 K, making NLTE models essential for accurate spectral analysis of hot stars.
• The NLTE model for Vega shows a good fit to observed FTS data without abundance fine-tuning, suggesting robustness of the NLTE treatment despite limited species coverage.