Detection of hot subdwarf binaries and sdB stars using machine learning methods and a large sample of Gaia XP spectra

TL;DR

This work uses Gaia XP spectra to systematically classify a large sample of hot subdwarfs, focusing on binarity and atmospheric properties. By applying UMAP to XP coefficients and training SOMs and CNN ensembles on both coefficient- and continuum-normalized spectra, the study identifies binary hot subdwarfs—primarily systems with main-sequence companions—while separating cool sdBs from hot/He-rich subdwarfs. The analysis uncovers a strong link between photometric/astrometric variability and binarity, but also reveals contamination by non-hot-sd populations (island A) and by cataclysmic variables in regions of the variability parameter space. The results demonstrate that Gaia XP data, coupled with dimensionality reduction and machine learning, illuminate the population structure and evolutionary paths of hot subdwarfs and guide follow-up spectroscopic validation.

Abstract

Hot subdwarfs (hot sds) are compact, evolved stars near the Extreme Horizontal Branch (EHB) and are key to understanding stellar evolution and the ultraviolet excess in galaxies. We extend our previous analysis of Gaia XP spectra of hot subdwarf stars to a much larger sample, enabling a comprehensive study of their physical and binary properties. Our goal is to identify patterns in Gaia XP spectra, investigate binarity, and assess the influence of parameters such as temperature, helium abundance, and variability. We analyse approximately 20000 hot subdwarf candidates selected from the literature, combining Gaia XP data with published parameters. We apply Uniform Manifold Approximation and Projection (UMAP) to the XP coefficients, which represent the Gaia XP spectra in a compact, feature-based form, to construct a similarity map. We then use self-organizing maps (SOMs) and convolutional neural networks (CNNs) to classify spectra as binaries or singles, and as cool and helium-poor or hot and helium-rich. The spectra are normalised using asymmetric least squares baseline fitting to emphasise individual spectral features. The BP-RP colour dominates the similarity map, with additional influence from temperature, helium abundance, and variability. Most binaries, identified via the Virtual Observatory SED Analyser (VOSA), cluster in two filaments linked to main sequence companions. CNN classification suggests a strong correlation between variability and binarity, with binary fractions exceeding 60 percent for active hot subdwarfs. Gaia XP spectra combined with dimensionality reduction and machine learning effectively reveal patterns in hot subdwarf properties. Our findings indicate that binarity and environmental density strongly shape the evolutionary paths of hot subdwarfs, and we identify possible contamination by main sequence and cataclysmic variable stars in the base sample.

Figures (17)

• Figure 1: UMAP of the 20061 sets of XP coefficients. The main body of the map is coloured in grey, separate regions are kept in black and labelled A, B, F1, and F2.
• Figure 2: Average of Gaia XP spectra of all objects in island A (black) and average XP spectrum of all other samples that are not is island A (grey). The insert panel highlights the region from 400 to 550 nm.
• Figure 3: Same UMAP as in Fig. \ref{['figure: UMAP_coeffs_groups']}, colour-coded by VOSA binarity. Our new data set has 2695 samples in common with set from Viscasillas2024.
• Figure 4: SOM map of the XP coefficients, with known hot sds colour-coded according to their binarity.
• Figure 5: UMAP of 20023 normalised XP spectra. All panels show the same UMAP, colour-coded by different parameters. Grey points represent spectra for which the respective parameter is not available.