Richard Hoppe (Germany)
hoppe @ mpia.de
Red giants as probes of nucleosynthesis and Galactic chemical evolution. Implications of 3D non-LTE spectral line formation
Large-scale stellar surveys, including SDSS-V, 4MOST, WEAVE, and PLATO, demand precise synthetic spectra for analyzing fundamental stellar parameters and chemical abundances. Owing to the computational complexity, previous studies often depended on 1D hydrostatic atmospheric models//combined with the simplifying assumption of local thermodynamic equilibrium (LTE). Besides LTE, these 1D models employed parametric approaches for convection (mixing-length theory, micro- and macro-turbulence). However, it has been shown that in the regime of FGK type stars, predictions from these models can significantly deviate from those derived using more sophisticated, physically realistic 3D model atmospheres and non-LTE radiative transfer. With our custom version of the MULTI-3D radiative transfer code, we have optimized the synthesis of spectra from 3D model atmospheres with non-LTE line formation and are now able to generate an extensive grid of stellar spectra with significantly reduced computational cost. This grid of 3D-non-LTE stellar spectra may be used to train various machine learning frameworks to determine stellar parameters from observations on-the-fly. Our goal is to develop a pipeline enabling the classification of metal-poor giants and subgiants from high-resolution 4MOST spectra. Within the scope of this thesis we will focus on abundances of a few key elements. C, O, Fe, Mg and Si abundances are of universal interest as they notably influence equations of state in almost all astrophysical applications while Sr, Y, Ba, Eu are important tracers of nucleosynthesis making these abundances valuable for galactic chemical evolution models. Additionally, we will publish the entire code used in the analysis making our version of MULTI-3D the first publicly available spectrum synthesis code for 3D stellar atmosphere models.
Supervisor: Maria Bergemann (MPIA)