Unveiling systematic biases in 1D LTE excitation-ionisation balance of Fe for FGK stars. A novel approach to determination of stellar parameters

TL;DR

This paper introduces a new spectroscopic analysis method for FGK stars that corrects for systematic biases in stellar parameters, improving accuracy for large stellar samples.

Contribution

It presents a novel approach combining NLTE corrections and Balmer line temperature scales to reduce biases in stellar parameter determination.

Findings

Significant biases up to 400 K in temperature were found using standard methods.

Systematic errors of 1.0 dex in surface gravity and 0.4 dex in metallicity were identified.

The new method improves the accuracy of stellar parameters for large surveys.

Abstract

We present a comprehensive analysis of different techniques available for the spectroscopic analysis of FGK stars, and provide a recommended methodology which efficiently estimates accurate stellar atmospheric parameters for large samples of stars. Our analysis includes a simultaneous equivalent width analysis of Fe I and Fe II spectral lines, and for the first time, utilises on-the-fly NLTE corrections of individual Fe I lines. We further investigate several temperature scales, finding that estimates from Balmer line measurements provide the most accurate effective temperatures at all metallicites. We apply our analysis to a large sample of both dwarf and giant stars selected from the RAVE survey. We then show that the difference between parameters determined by our method and that by standard 1D LTE excitation-ionisation balance of Fe reveals substantial systematic biases: up to 400 K in effective temperature, 1.0 dex in surface gravity, and 0.4 dex in metallicity for stars with [Fe/H] ~ -2.5. This has large implications for the study of the stellar populations in the Milky Way.