J/AJ/170/158 BHB identification in LAMOST DR10 with deep learning (Zhang+, 2025)
An advanced deep learning model for identifying blue horizontal-Branch stars
from LAMOST DR10.
Zhang Y., Bu Y., Zhang J., Wang Ke, Wu H., Zhang M., Li S., Sun J.,
Kong X., Yi Z., Liu M.
<Astron. J., 170, 158 (2025)>
=2025AJ....170..158Z 2025AJ....170..158Z
ADC_Keywords: Stars, horizontal branch; Spectroscopy; Photometry; Optical;
Models
Keywords: Astrostatistics ; Astronomy data analysis ; Horizontal branch stars ;
Convolutional neural networks
Abstract:
Blue horizontal-branch (BHB) stars are ideal tracers for studying the
kinematics and structural properties of the Milky Way. With massive
spectral data provided by the Large Sky Area Multi-Object Fiber
Spectroscopic Telescope (LAMOST), we aim to identify more potential
BHB stars using machine learning methods. In this study, we propose
BHBNet, an advanced two-stage deep learning model integrating multiple
techniques. By implementing Bayesian inference, it not only provides
classification results but also quantifies uncertainty. In stage 1, a
six-class classification model was constructed to initially identify
BHB candidates, achieving a precision of 95.43% on the test set. In
stage 2, a binary classification model constructed through the
transfer learning method was employed to further refine the
candidates, achieving a precision of 98.36% on the test set.
Subsequently, by performing a two-stage search in LAMOST
low-resolution survey DR10, we identified 6792 candidates.
Nevertheless, since the completeness of this search result has not
been assessed, these samples may not be adequate for statistical
studies of the BHB population. We analyzed candidate properties
including color, absolute magnitude, and spatial distribution, while
estimating their atmospheric parameters. Eventually, by fitting Balmer
line profiles, we identified 1605 new BHB stars compared to the
previous studies by X.-X. Xue et al. and J. J. Vickers et al. Our
study emphasizes the potential and effectiveness of using machine
learning methods in identifying and analyzing BHB stars.
Description:
LAMOST (DR11, V/162), also known as the Guo Shou Jing Telescope, is a
large-aperture and wide-field telescope independently designed by
Chinese astronomers. This telescope can capture 4000 spectra in a
single exposure, with a limiting magnitude of r=19. The 10th data
release of the LAMOST survey has published over 20 million spectral
data. This study utilizes the spectra from LAMOST-LRS DR10, with a
wavelength coverage of 3700-9000Å and R∼1800, which have been flux
and wavelength calibrated.
To design and evaluate the classifier, we constructed two sample sets:
a positive sample set comprising BHB stars and a negative sample set
containing non-BHB stars. For the positive sample set, we
cross-matched the BHB catalog provided by J. J. Vickers et al. (2021)
with the LAMOST-LRS DR10 catalog.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table6.dat 181 6792 The Blue horizontal-branch (BHB) catalog
--------------------------------------------------------------------------------
See also:
I/355 : Gaia DR3 Part 1. Main source (Gaia Collaboration, 2022)
V/162 : LAMOST DR11 catalogs (Luo+, 2026)
II/306 : The SDSS Photometric Catalog, Release 8 (Adelman-McCarthy+, 2011)
II/358 : SkyMapper Southern Sky Survey. DR1.1 (Wolf+, 2018)
II/371 : The Dark Energy Survey (DES): Data Release 2 (Abbott+, 2021)
III/286 : APOGEE-2 DR17 final allStar catalog (Abdurro'uf+, 2022)
J/AJ/103/267 : Spectroscopy of hot stars in galactic halo (Beers+ 1992)
J/AJ/127/899 : Blue HB stars in SDSS (Sirko+, 2004)
J/ApJ/684/1143 : BHB candidates in the Milky Way (Xue+, 2008)
J/A+A/522/A88 : Photometric identification of BHB stars (Smith+, 2010)
J/ApJ/731/119 : BHB candidates in Sagittarius stream (Ruhland+, 2011)
J/ApJ/738/79 : SDSS-DR8 BHB stars in the Milky Way's halo (Xue+, 2011)
J/ApJ/750/99 : The Pan-STARRS1 photometric system (Tonry+, 2012)
J/A+A/600/A50 : Catalog of hot subdwarf stars (Geier+, 2017)
J/MNRAS/488/2892 : Gaia DR2 extremely low-mass WD candidates (Pelisoli+, 2019)
J/ApJ/886/154 : Sgr stream K- & M-giants and BHB stars (Yang+, 2019)
J/A+A/654/A107 : Catalogues of Blue Horizontal Branch Stars (Culpan+, 2021)
J/ApJS/256/28 : Hot subdwarf stars with Gaia DR2 & LAMOST DR7 data (Luo+, 2021)
J/ApJS/256/14 : RVs from LAMOST MRS DR7 stellar spectra (Zhang+, 2021)
J/ApJS/259/5 : Hot subdwarf candidates from LAMOST DR7 (Tan+, 2022)
J/A+A/662/A66 : Hot stars from LAMOST DR6 (Xiang+, 2022)
J/ApJS/276/53 : White dwarfs from machine learning (Zhang+, 2025)
Byte-by-byte Description of file: table6.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 10 I10 --- ObsID LAMOST-LRS DR10 (V/162) unique spectrum
identifier
12- 26 A15 --- uID LAMOST DR10 (V/162) unique source identifier
28- 37 F10.6 deg RAdeg Right Ascension in decimal degrees (J2000)
39- 47 F9.6 deg DEdeg Declination in decimal degrees (J2000)
49- 54 F6.2 --- SNR [10/740] LAMOST DR10 g-band Signal-to-Noise
56- 74 I19 --- GaiaDR3 ? Gaia DR3 (I/355) unique source identifier
76- 80 F5.2 mas plx [-4.66/6.19]? Gaia DR3 parallax
82- 87 F6.2 --- RPlx [-5.51/181.6]? Gaia DR3 parallax_over_error
value (plxErr)
89- 93 F5.2 mag Gmag [6.2/20.8]? Apparent Gaia DR3 G-band
magnitude
95- 99 F5.2 mag BP-RP [-0.11/2.62]? Gaia DR3 Blue-Reb color
101- 105 F5.2 mag GMag [-1.76/10.14]? Absolute Gaia DR3 G-band
magnitude
107- 111 F5.2 kpc Dist [0.16/10.24]? Distance from Gaia DR3
parallax (Dis)
113- 117 F5.2 kpc Rgc [1.03/16.61]? Galactocentric distance (R)
119- 123 F5.2 kpc Zgc [-9.16/8.94]? Vertical distance to Galactic
disk middle (Z)
125- 129 I5 K Teff [7005/35267] Effective temperature
131- 135 F5.2 [cm/s2] logg [-1.73/5.64] Log surface gravity
137- 141 F5.2 [-] [Fe/H] [-9.0/1.4] Metallicity
143- 181 A39 --- Source Source information (1)
--------------------------------------------------------------------------------
Note (1): Source as follows:
Candidates = 2475 occurrences
New = 1605 occurrences
Vickers et al. (2021) = 2021ApJ...912...32V 2021ApJ...912...32V, 2668 occurrences
Xue et al. (2011) = 2011ApJ...738...79X 2011ApJ...738...79X, J/ApJ/738/79, 217 occurrences
--------------------------------------------------------------------------------
History:
From electronic version of the journal
License: CC-BY-4.0
(End) Prepared by [AAS], Robin Leichtnam [CDS] 01-Jul-2026