Image Details

Choose export citation format:

Evaluating Classifications of Extremely Metal-poor Candidates Selected from Gaia XP Spectra

  • Authors: Riley Thai, Andrew R. Casey, Alexander P. Ji, Vedant Chandra, Hans-Walter Rix

Riley Thai et al 2026 The Astronomical Journal 171 .

  • Provider: AAS Journals

Caption: Figure 6.

Other catalogs in the literature can robustly model [M/H] to low metallicities using different approaches. Shown is a comparison of modeled metallicities [M/H], augmented with both EMP and UMP stars from the SAGA database to different catalogs of metallicities available in the literature. These are as follows: N. F. Martin et al. (2024), using synthesized Ca H and K metallicities from Gaia XP spectra; Y. Yao et al. (2024), using a dereddened set of XP coefficients and a larger training sample of metal-poor stars (their probability of being not VMP or lower, [Fe/H] > −2, is plotted); A. Khalatyan et al. (2024), using XGBoost on a dereddened set of Gaia XP coefficients and a substantially larger training set that includes hotter stars; L. Yang et al. (2025), using a cost-sensitive neural network on a large sample of metal-poor stars from the PASTEL/SAGA catalogs. Where reported, model errors are plotted in the upper left corner. See text for details.

Other Images in This Article

Show More

Copyright and Terms & Conditions

Additional terms of reuse