Pith. sign in

REVIEW 3 major objections 5 minor 52 references

The Stellar Abundances and Galactic Evolution Survey (SAGES). II. Machine Learning-Based Stellar parameters for 21 million stars from the First Data Release

T0 review · 3 major / 5 minor · reviewed 2026-08-09 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read This paper claims SAGES uv photometry plus Gaia colors, run through a random forest trained on spectroscopy, recovers metallicity, surface gravity, and temperature for 21 million Northern-sky stars at 0.09-dex, 0.12-dex, and 70-K precision.

desk verdict A useful 21M-star northern photometric catalog with a clean ML pipeline, but the accuracy claims are undercut by an unaddressed 0.6 dex offset for one open cluster and an unexplained calibration shift. read the letter →

arxiv 2502.03548 v1 pith:JXA3FEC6 submitted 2025-02-05 astro-ph.SR astro-ph.GA

Hongrui Gu , Zhou Fan , Gang Zhao , Yang Huang , Timothy C. Beers , Wei Wang , Jie Zheng , Jingkun Zhao
show 9 more authors
Chun Li Yuqin Chen Haibo Yuan Haining Li Kefeng Tan Yihan Song Ali Luo Nan Song Yujuan Liu
This is my paper · ORCID
classification astro-ph.SRastro-ph.GA
keywords SAGESsurveystellarparametersphotometricmetallicityrandomforestmetal-poorstarsGaiaEDR3machinelearningMilkyWaystructure
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper claims that stellar parameters can be extracted from photometry alone for much of the Northern sky, without taking spectra. The authors combine SAGES medium-band u and v photometry, bands that straddle the Balmer jump and sample the Ca II H and K lines, with Gaia EDR3 colors and parallaxes, and train random-forest models on spectroscopic labels from LAMOST DR10, validated against APOGEE and several other surveys. The result is a catalog of $[\mathrm{Fe/H}]$, $\log g$, and $T_{\mathrm{eff}}$ for 21,071,305 stars, split into 19,663,040 dwarfs and 1,408,265 giants, with claimed precisions of about 0.09 dex, 0.12 dex, and 70 K. This matters because a photometric catalog at these precisions is roughly two orders of magnitude larger than current spectroscopic samples, and it directly feeds metal-poor star selection and Milky Way structure studies.

What carries the argument

The carrying mechanism is the random forest, an ensemble of decision trees that split on the Gini index, configured with 300 trees, all features per split, and a maximum depth of 17, and trained as five separate models: $[\mathrm{Fe/H}]$ for dwarfs, $[\mathrm{Fe/H}]$ for giants, $T_{\mathrm{eff}}$ for dwarfs, $T_{\mathrm{eff}}$ for giants, and one unified $\log g$ model for both classes. The physical carrier is the SAGES filter design: the u and v bands straddle the Balmer jump, which is sensitive to surface gravity, while the v band sits on the Ca II H and K lines, which are sensitive to metallicity, and the Gaia Bp/Rp colors together with the parallax-based absolute G magnitude carry temperature and the dwarf/giant separation. Model inputs are extinction-corrected magnitudes, reddening-corrected colors, parallax, absolute G magnitude, and $\mathrm{E}(B-V)$, with the SFD98 map providing the extinction corrections. Stars redder than $(Bp-Rp)_0 = 1.598$ are excluded from the final catalog because the forest extrapolates poorly at low temperature.

What would settle it

High-resolution spectroscopy of a few hundred catalog stars in the metal-poor regime would settle the scale question: if the photometric $[\mathrm{Fe/H}]$ values below $[\mathrm{Fe/H}] \approx -2$ deviate from the spectroscopic values by more than the claimed 0.2–0.3 dex, or if the residuals trend with metallicity instead of staying flat, the single-offset scale inheritance fails. A faster check is to cross-match the nine literature metal-poor samples against LAMOST directly and test whether one constant +0.427 offset actually reconciles them.

Watch

Extended reading notes

Core claim

The central discovery claim is that the SAGES u and v bands combined with Gaia EDR3 broad-band colors and parallaxes carry enough information for a random forest to recover $[\mathrm{Fe/H}]$, $\log g$, and $T_{\mathrm{eff}}$ at precisions of 0.09 dex, 0.12 dex, and 70 K, for both dwarf and giant stars. The metallicity scale is set by the training labels: LAMOST DR10 parameters for the bulk of the sample, supplemented at the metal-poor end by PASTEL and RAVE, with nine high-resolution literature samples used as a low-metallicity test after a single +0.427 dex offset is applied to place them on the LAMOST scale. Adding 2MASS, WISE, and GALEX photometry improves precision for a subsample of 2,191,452 stars, but the authors retain the larger SAGES-plus-Gaia sample as the main catalog because the richer dataset suffers stronger selection effects. The paper further claims the precision degrades to only 0.2–0.3 dex below $[\mathrm{Fe/H}] = -2$, and that cluster and external-catalog comparisons are consistent with the quoted dispersions.

Load-bearing premise

The catalog's absolute accuracy is inherited whole from the LAMOST DR10 spectroscopic parameter scale used as training labels, together with a single +0.427 dex offset applied to nine independent metal-poor samples; if that scale is biased, or if one constant cannot reconcile the literature scales, every $[\mathrm{Fe/H}]$ in the final catalog shifts systematically, and the stated 0.09 dex is an agreement with APOGEE rather than an absolute calibration.

Editorial extensions

If this is right

  • The catalog gives photometric $[\mathrm{Fe/H}]$, $\log g$, and $T_{\mathrm{eff}}$ for 21,071,305 Northern-sky stars, 19,663,040 dwarfs and 1,408,265 giants, at claimed precisions of about 0.09 dex, 0.12 dex, and 70 K.
  • Dispersion grows to 0.2–0.3 dex only for $[\mathrm{Fe/H}] < -2$, so the catalog can supply very and extremely metal-poor candidates without new spectroscopy.
  • The five-model setup covers giants and dwarfs with the same pipeline, so the method applies across the color-magnitude diagram rather than only to FGK dwarfs.
  • Because random forests accept arbitrary feature sets, the same models can absorb the upcoming SAGES DDO51 and H-alpha bands, which the paper points out should improve $\log g$ in particular.
  • In the main catalog, about 0.03% of dwarfs and 0.21% of giants sit at $[\mathrm{Fe/H}] \leq -2$, quantifying the metal-poor candidate yield the catalog offers.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A corollary the paper leaves implicit is that the catalog extrapolates the LAMOST DR10 metallicity scale to roughly twenty times more stars, so a future recalibration of LAMOST would propagate through nearly every entry; the $[\mathrm{Fe/H}]$ values are best read as precise but anchored to that one external scale.
  • The single +0.427 dex offset is the likeliest place the metal-poor tail could bend: if the offset is metallicity-dependent rather than constant, the catalog's $[\mathrm{Fe/H}] < -2$ statistics would be compressed or stretched while near-solar values stay correct.
  • The ~50 K systematic temperature offset against the H23 catalog hints that part of the error budget lives in the SAGES DR1 photometry itself, which would make the planned DDO51 and H-alpha bands not just precision upgrades but a possible handle on those systematics.
  • The same filter logic, a Balmer-jump straddle plus a Ca II H and K band, is shared with other recent medium-band surveys, so this random-forest recipe is a transferable template for turning their photometry into stellar parameters.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper presents a machine-learning pipeline, based on the random forest algorithm, that estimates effective temperature, surface gravity, and metallicity for approximately 21 million stars by combining SAGES DR1 u/v photometry with Gaia EDR3 photometry and astrometry. A second catalog of about 2.2 million stars additionally incorporates 2MASS, WISE, and GALEX photometry. The models are trained on LAMOST DR10 parameters and tested on APOGEE, with additional external checks against the H23 and A23 catalogs and against four star clusters. The claimed precisions are about 0.09 dex in [Fe/H], 0.12 dex in log g, and 70 K in Teff. The manuscript also applies a +0.427 dex correction to the metallicities of nine metal-poor literature samples used as a low-metallicity test set. The paper concludes that the resulting catalogs are suitable for metal-poor star selection and Milky Way structure studies.

Significance. If the stated precisions are accurate, the catalog would be a valuable resource for Galactic archaeology and for selecting metal-poor and extremely metal-poor candidates for follow-up spectroscopy. The validation design is sound in outline: a separate APOGEE test set, comparisons with two published catalogs, and cluster-based checks go beyond many comparable photometric surveys. The strongest features are the use of a medium-band filter set with metallicity sensitivity and the explicit recognition that train/test separation across different spectroscopic surveys is necessary. However, the absolute calibration ultimately rests on the LAMOST scale, and the paper contains two validation issues that directly affect the accuracy claim: the unexplained +0.427 dex offset and an unaddressed ~0.6 dex discrepancy for the open cluster NGC 2420. These issues do not invalidate the catalog's usefulness as a relative indicator, but they must be resolved or clearly quantified before the absolute accuracy claims in the abstract and Section 3.3 can be accepted.

major comments (3)
  1. [Section 2.3] The statement that 'a correction of 0.427 in [Fe/H] is applied to the data from 9 papers' is not accompanied by any explanation of how this number was derived. If the offset was tuned to minimize residuals between the photometric predictions and the literature values for the same stars, then the subsequent low-metallicity comparison in Section 3.5 is not an independent test. Please provide the derivation, the paper-by-paper offsets and their scatter, and justify why a single additive constant is sufficient to reconcile the scales of nine different studies.
  2. [Section 3.5, Figure 11] Figure 11 shows that for the metal-rich open cluster NGC 2420 the catalog mean metallicity is -0.339 ± 0.088, while the quoted Cantat-Gaudin et al. (2018) value is +0.279, a discrepancy of about -0.6 dex. The paper states that the metal-poor cluster offsets are consistent with the test set but does not mention NGC 2420 at all; this is a direct test of the absolute metallicity scale in the regime where Section 3.3 claims accuracy better than 0.1 dex. The authors should either explain this offset (e.g., member contamination, metallicity scale differences, or a systematic in the model) or amend the accuracy claims in the abstract and Section 3.3 to exclude this regime.
  3. [Section 3.3] The quoted precision of 0.09 dex refers to the scatter against the APOGEE test set. Because the model is trained on LAMOST DR10 labels, the absolute zero point of the catalog is inherited from LAMOST, and any metallicity- or temperature-dependent bias in those labels propagates directly into the final catalog. The paper should explicitly state that the APOGEE comparison measures precision and relative accuracy, not absolute calibration against an independent standard, and should discuss the extent to which the external catalog and cluster comparisons constrain the LAMOST scale.
minor comments (5)
  1. [Table 3] There are typos in the table: 'parallex' should be 'parallax' and 'Uncentainty' should be 'Uncertainty' in the magnitude and proper-motion entries.
  2. [Section 2.2] The phrase 'the most metal-poor star reliably estimated by most pipelines in DR10 is only be about [Fe/H] ∼ −2.5' contains a grammatical error ('is only be about') and could be rephrased.
  3. [Figure 12] The histogram bin labels such as '1 < [Fe/H] 0' should be written with clear inequalities (e.g., '-1 < [Fe/H] ≤ 0') to avoid ambiguity.
  4. [Section 4] The sentence about removing low-temperature stars with (Bp − Rp)0 > 1.598 would benefit from a brief justification of this cut and a statement of how many stars were discarded; the point that random forest extrapolates poorly is made but the threshold choice is not discussed.
  5. [Section 3.4] The comparison with H23 shows a ~50 K systematic difference in Teff that the authors attribute to the training set; this is worth a short discussion of whether it also affects the log g or [Fe/H] zero points.

Circularity Check

1 steps flagged · score 2.0 of 10

Minor scale-calibration caveat at low metallicity, but the central 21-million-star parameter derivation is not circular: it is an empirical photometric-to-spectroscopic mapping validated against independent APOGEE labels.

  1. other [Section 2.3, repeated in Section 2.5; used in the low-metallicity validation in Sections 3.3 and 3.5]
    "Considering the systematic differences in the parameters provided by different datasets, a correction of 0.427 in [Fe/H] is applied to the data from 9 papers to account for these systematic differences."

    The nine-paper test labels are transformed to [Fe/H]_lit + 0.427 before comparison with model predictions, placing them on the LAMOST training scale. The paper's claim that the metal-poor systematic errors are consistent with the test set is therefore a statement about agreement with scale-shifted labels, not an independently calibrated absolute scale. If 0.427 was estimated from these same datasets, the mean low-metallicity residual is partly forced toward zero by construction. This compromises only the low-metallicity validation; the headline 0.09 dex precision rests on the APOGEE test set and remains an independent empirical measurement.

full rationale

The central derivation is a supervised random forest mapping from SAGES DR1 + Gaia EDR3 photometry to stellar parameters, trained on LAMOST labels and tested on APOGEE, which is an independent spectroscopic survey. The reported 0.09 dex [Fe/H] precision is an empirical generalization error from that APOGEE test set (Figures 6-8), not a fitted re-description of the training labels. The catalog values are not defined in terms of the quantities they predict. External checks with Gaia XP spectra (A23), star clusters (H10/C18), and the independent APOGEE test set provide non-circular support; the H23 comparison uses the same SAGES survey and overlapping authors and is therefore not an independent anchor, but it is not load-bearing for the main claims. The only calibration concern is the +0.427 dex offset applied to nine low-metallicity literature samples before using them as test labels (Sections 2.3 and 2.5). Because the offset is applied without derivation, the low-metallicity validation is partly dependent on a scale choice, and the NGC 2420 cluster comparison shows an unaddressed ~0.6 dex discrepancy that warrants care. However, this does not make the 21-million-star catalog derivation circular; it is a scale-calibration caveat. No self-citation chain is used to establish the central result, and no ansatz is smuggled in via citation. Overall circularity is minor.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

No new physical entities are postulated. The central mapping is empirical, so the ledger is mostly survey-analysis assumptions: the accuracy of spectroscopic labels, the SFD98 reddening correction, and the parallax-based dwarf/giant split. The one fitted value that directly shapes the headline precision is the +0.427 dex offset applied to the nine metal-poor validation samples (Section 2.3). The random-forest hyperparameters and the cool-star color cut are hand-chosen inputs that define the working range of the catalog.

free parameters (3)
  • Systematic [Fe/H] offset for nine metal-poor literature samples = +0.427 dex
    Applied to the nine collection papers' [Fe/H] values in Section 2.3 to align their scale with LAMOST; the value is stated without derivation, and the low-metallicity validation results depend on it.
  • Random forest hyperparameters = n_estimators = 300, max_depth = 17, max_features = None
    Chosen via out-of-bag score on the training set (Section 2.4, Figure 4). This is standard model selection rather than fitting to the test claim, but it is a hand-chosen input to the final model.
  • Cool-star exclusion cut = (Bp - Rp)0 > 1.598 removed
    Section 4: low-temperature extrapolation was unsatisfactory, so all three parameters are flagged -9999 for cooler stars; this post-hoc choice defines the effective validity range of the catalog.
assumptions (5)
  • domain assumption Spectroscopic stellar parameters from LAMOST DR10, APOGEE DR17, RAVE DR5, and PASTEL are accurate on a common scale sufficient for training labels.
    Invoked in Sections 2.2 to 2.3; the photometric mapping's absolute accuracy inherits the label scale, including any metallicity-dependent bias.
  • domain assumption The SFD98 reddening map and the adopted reddening coefficients (Table 1) correctly de-redden the SAGES, Gaia, 2MASS, WISE, and GALEX photometry.
    Section 2.1; incorrect extinction correction would bias the colors and therefore all derived parameters, especially toward the Galactic plane.
  • domain assumption Gaia parallax quality (parallax error larger than the parallax indicates a giant) reliably separates dwarfs from giants.
    Section 2.1 and Table 2; the dwarf and giant models are trained separately, so misclassification feeds label noise into both.
  • domain assumption The photometric color-to-parameter relation is single-valued over the training manifold, and random forest extrapolation outside it fails only in the cool-star regime.
    Sections 2.4 and 4; the paper's solution for cool stars is a color cut rather than a model of extrapolation failure.
  • ad hoc to paper A single additive +0.427 dex shift reconciles the parameter scales of nine independent metal-poor papers with the LAMOST scale.
    Section 2.3; no justification or fit diagnostics are given, and the metal-poor validation accuracy depends on this offset.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The Stellar Abundances and Galactic Evolution Survey (SAGES). II. Machine Learning-Based Stellar parameters for 21 million stars from the First Data Release." pith.science (2026). https://pith.science/paper/JXA3FEC6

@misc{pith2026250203548,
  author       = {Pith},
  title        = {Pith review of: The Stellar Abundances and Galactic Evolution Survey (SAGES). II. Machine Learning-Based Stellar parameters for 21 million stars from the First Data Release},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JXA3FEC6}},
  note         = {Machine review of arXiv:2502.03548}
}
abstract

Stellar parameters for large samples of stars play a crucial role in constraining the nature of stars and stellar populations in the Galaxy. An increasing number of medium-band photometric surveys are presently used in estimating stellar parameters. In this study, we present a machine-learning approach to derive estimates of stellar parameters, including [Fe/H], logg, and Teff, based on a combination of medium-band and broad-band photometric observations. Our analysis employs data primarily sourced from the SAGE Survey , which aims to observe much of the Northern Hemisphere. We combine the $uv$-band data from SAGES DR1 with photometric and astrometric data from Gaia EDR3, and apply the random forest method to estimate stellar parameters for approximately 21 million stars. We are able to obtain precisions of 0.09 dex for [Fe/H], 0.12 dex for logg, and 70 K for Teff. Furthermore, by incorporating 2MASS and WISE infrared photometric and GALEX ultraviolet data, we are able to achieve even higher precision estimates for over 2.2 million stars. These results are applicable to both giant and dwarf stars. Building upon this mapping, we construct a foundational dataset for research on metal-poor stars, the structure of the Milky Way, and beyond. With the forthcoming release of additional bands from SAGE Survey such DDO51 and H-alpha, this versatile machine learning approach is poised to play an important role in upcoming surveys featuring expanded filter sets

Figures

Figures reproduced from arXiv: 2502.03548 by the authors.

Figure 1
Figure 1. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Left two panels: Number-density distribution of the training samples for Data Set 1 in the color-absolute magnitude diagram, and the histogram of [Fe/H] from different spectral survey catalogs. Right two panels: Same to left two panels but from test set [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Similar to [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Left panel: The Out-of-Bag (OOB) scores for different combinations of hyperparameters are shown, with different colors representing different max features values, and the x-axis representing different n estimators values. The y-axis represents 1 − OOB score, so a small…
Figure 5
Figure 5. Figure 5: Flow chart of this work [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Comparisons between the spectroscopic and photometric stellar parameters for the test set from Data Set 1, for 5 models (dwarf [Fe/H], giant [Fe/H], dwarf Teff , giant Teff , log g). The lower portion of each panel shows the number density distribution of the differenc…
Figure 7
Figure 7. Figure 7: Similar to [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]
Figure 8
Figure 8. Figure 8: Systematic errors and standard deviations for three stellar parameters ([Fe/H], Teff , log g), as a function of their values. Different colors represent different test sets; the different shapes of the points represent different types of stars (triangles for dwarfs, sq…
Figure 9
Figure 9. Figure 9: A comparison of the performance of the H23 catalog and our catalog for four models with two parameters. The lower half of each panel shows the distribution of the differences between our model and the H23 model, based on the parameters provided by the H23 model, along …
Figure 10
Figure 10. Figure 10: Similar to [PITH_FULL_IMAGE:figures/full_fig_p008_10.png]
Figure 11
Figure 11. Figure 11: This figure presents the metallicity histograms from this work. The top two panels show the open clusters NGC 2420 and NGC 2682, and the bottom two panels display the globular clusters NGC 6934 and NGC 7089. The red dashed lines represent the best-fit Gaussian distrib…
Figure 12
Figure 12. Figure 12: Left panel: The photometric-metallicity distributions of dwarf (blue) and giant (yellow) stars in Data Set 1. The number of dwarf and giant stars in each bin of metallicity are also given. Right panel: Similar to left panel, but using data from Data Set 2 [PITH_FULL_…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

52 extracted references · 13 canonical work pages

  1. [1]

    2023, VizieR Online Data Catalog: APOGEE-2 DR17 final allStar catalog (Abdurro’uf+, 2022), VizieR On-line Data Catalog: III/286

    Abdurro’Uf, & et al. 2023, VizieR Online Data Catalog: APOGEE-2 DR17 final allStar catalog (Abdurro’uf+, 2022), VizieR On-line Data Catalog: III/286. Originally published in: 2022ApJS..259...35A

    work page 2023

  2. [2]

    An, D., & Beers, T. C. 2020, ApJ, 897, 39, doi: 10.3847/1538-4357/ab8d39

    work page doi:10.3847/1538-4357/ab8d39 2020

  3. [3]

    2023, ApJS, 267, 8, doi: 10.3847/1538-4365/acd53e

    Andrae, R., Rix, H.-W., & Chandra, V . 2023, ApJS, 267, 8, doi: 10.3847/1538-4365/acd53e

    doi:10.3847/1538-4365/acd53e 2023

  4. [4]

    C., Lee, Y

    Aoki, W., Beers, T. C., Lee, Y . S., et al. 2013, AJ, 145, 13, doi: 10.1088/0004-6256/145/1/13

    doi:10.1088/0004-6256/145/1/13 2013

  5. [5]

    2019, AJ, 158, 93, doi: 10.3847/1538-3881/ab3048

    Bai, Y ., Liu, J., Bai, Z., Wang, S., & Fan, D. 2019, AJ, 158, 93, doi: 10.3847/1538-3881/ab3048

    doi:10.3847/1538-3881/ab3048 2019

  6. [6]

    2014, VizieR Online Data Catalog: GALEX-GR6/7 data release (Bianchi+ 2014), VizieR On-line Data Catalog: II/335

    Bianchi, L., Conti, A., & Shiao, B. 2014, VizieR Online Data Catalog: GALEX-GR6/7 data release (Bianchi+ 2014), VizieR On-line Data Catalog: II/335. Originally published in: 2014AdSpR..53..900B

    work page 2014

  7. [7]

    2009, A&A, 501, 519, doi: 10.1051/0004-6361/200810610

    Bonifacio, P., Spite, M., Cayrel, R., et al. 2009, A&A, 501, 519, doi: 10.1051/0004-6361/200810610

    doi:10.1051/0004-6361/200810610 2009

  8. [8]

    2021, A&A, 653, A31, doi: 10.1051/0004-6361/202038841

    Bonoli, S., Mar´ın-Franch, A., Varela, J., et al. 2021, A&A, 653, A31, doi: 10.1051/0004-6361/202038841

    doi:10.1051/0004-6361/202038841 2021

Show all 52 references
  1. [9]

    2001, Machine Learning, 45, 5, doi: 10.1023/A:1010933404324

    Breiman, L. 2001, Machine Learning, 45, 5, doi: 10.1023/A:1010933404324

    2001 doi

  2. [10]

    2018, A&A, 618, A93, doi: 10.1051/0004-6361/201833476

    Cantat-Gaudin, T., Jordi, C., Vallenari, A., et al. 2018, A&A, 618, A93, doi: 10.1051/0004-6361/201833476

    2018 doi

  3. [11]

    J., Moles, M., Crist´obal-Hornillos, D., et al

    Cenarro, A. J., Moles, M., Crist´obal-Hornillos, D., et al. 2019, A&A, 622, A176, doi: 10.1051/0004-6361/201833036

    2019 doi

  4. [12]

    K., et al

    Chiti, A., Frebel, A., Mardini, M. K., et al. 2021, ApJS, 254, 31, doi: 10.3847/1538-4365/abf73d

    2021 doi

  5. [13]

    L., Barnes, J

    Crawford, D. L., Barnes, J. V ., & Golson, J. C. 1970, AJ, 75, 624, doi: 10.1086/110996

    1970 doi

  6. [14]

    2012, Research in Astronomy and Astrophysics, 12, 1197, doi: 10.1088/1674-4527/12/9/003

    Cui, X.-Q., Zhao, Y .-H., Chu, Y .-Q., et al. 2012, Research in Astronomy and Astrophysics, 12, 1197, doi: 10.1088/1674-4527/12/9/003

    2012 doi

  7. [15]

    M., Skrutskie, M

    Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, 2MASS All Sky Catalog of point sources

    2003

  8. [16]

    M., Wright, E

    Cutri, R. M., Wright, E. L., Conrow, T., et al. 2021, VizieR Online Data

    2021

  9. [17]

    2020, ApJ, 898, 150, doi: 10.3847/1538-4357/ab9d1a

    Ezzeddine, R., Rasmussen, K., Frebel, A., et al. 2020, ApJ, 898, 150, doi: 10.3847/1538-4357/ab9d1a

    2020 doi

  10. [18]

    2023, ApJS, 268, 9, doi: 10.3847/1538-4365/ace04a Franc ¸ois, P., Caffau, E., Bonifacio, P., et al

    Fan, Z., Zhao, G., Wang, W., et al. 2023, ApJS, 268, 9, doi: 10.3847/1538-4365/ace04a Franc ¸ois, P., Caffau, E., Bonifacio, P., et al. 2018, A&A, 620, A187, doi: 10.1051/0004-6361/201834375 Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al. 2021, A&A, 649, A1, doi: 10...

    2023 doi

  11. [19]

    A., Daflon, S., Placco, V

    Galarza, C. A., Daflon, S., Placco, V . M., et al. 2022, A&A, 657, A35, doi: 10.1051/0004-6361/202141717

    2022 doi

  12. [20]

    Harris, W. E. 2010, arXiv e-prints, arXiv:1012.3224, doi: 10.48550/arXiv.1012.3224

    2010 doi arXiv

  13. [21]

    1998, A&AS, 129, 431, doi: 10.1051/aas:1998195

    Hauck, B., & Mermilliod, M. 1998, A&AS, 129, 431, doi: 10.1051/aas:1998195

    1998 doi

  14. [22]

    Q., Yuan, H

    Huang, Y ., Chen, B. Q., Yuan, H. B., et al. 2019, ApJS, 243, 7, doi: 10.3847/1538-4365/ab1f72

    2019 doi

  15. [23]

    C., Wolf, C., et al

    Huang, Y ., Beers, T. C., Wolf, C., et al. 2022, ApJ, 925, 164, doi: 10.3847/1538-4357/ac21cb

    2022 doi

  16. [24]

    C., Yuan, H., et al

    Huang, Y ., Beers, T. C., Yuan, H., et al. 2023, ApJ, 957, 65, doi: 10.3847/1538-4357/ace628 Ivezi´c, ˇZ., Sesar, B., Juri´c, M., et al. 2008, ApJ, 684, 287, doi: 10.1086/589678

    2023 doi

  17. [25]

    2007, in Astronomical Society of the Pacific Conference Series, V ol

    Keller, S., Bessell, M., Schmidt, B., & Francis, P. 2007, in Astronomical Society of the Pacific Conference Series, V ol. 364, The Future of

    2007

  18. [26]

    F., Ardizzone, L., Klessen, R., et al

    Ksoll, V . F., Ardizzone, L., Klessen, R., et al. 2020, MNRAS, 499, 5447, doi: 10.1093/mnras/staa2931

    2020 doi

  19. [27]

    2017, AJ, 153, 75, doi: 10.3847/1538-3881/153/2/75

    Kunder, A., Kordopatis, G., Steinmetz, M., et al. 2017, AJ, 153, 75, doi: 10.3847/1538-3881/153/2/75

    2017 doi

  20. [28]

    K., Bolte, M., Johnson, J

    Lai, D. K., Bolte, M., Johnson, J. A., & Lucatello, S. 2004, in American Astronomical Society Meeting Abstracts, V ol. 205, American Astronomical Society Meeting Abstracts, 52.10

    2004

  21. [29]

    2018, ApJS, 238, 16, doi: 10.3847/1538-4365/aada4a

    Li, H., Tan, K., & Zhao, G. 2018, ApJS, 238, 16, doi: 10.3847/1538-4365/aada4a

    2018 doi

  22. [30]

    C., et al

    Limberg, G., Rossi, S., Beers, T. C., et al. 2021, ApJ, 907, 10, doi: 10.3847/1538-4357/abcb87

    2021 doi

  23. [31]

    2024, ApJS, 271, 26, doi: 10.3847/1538-4365/ad1eea

    Lu, X., Yuan, H., Xu, S., et al. 2024, ApJS, 271, 26, doi: 10.3847/1538-4365/ad1eea

    2024 doi

  24. [32]

    R., Schiavon, R

    Majewski, S. R., Schiavon, R. P., Frinchaboy, P. M., et al. 2017, AJ, 154, 94, doi: 10.3847/1538-3881/aa784d

    2017 doi

  25. [33]

    F., Starkenburg, E., Yuan, Z., et al

    Martin, N. F., Starkenburg, E., Yuan, Z., et al. 2023, arXiv e-prints, arXiv:2308.01344, doi: 10.48550/arXiv.2308.01344 Mendes de Oliveira, C., Ribeiro, T., Schoenell, W., et al. 2019, MNRAS, 489, 241, doi: 10.1093/mnras/stz1985 Nordstr¨om, B., Andersen, J., Holmberg, J., et a...

    2023 doi arXiv

  26. [34]

    M., Lee, Y

    Rockosi, C. M., Lee, Y . S., Morrison, H. L., et al. 2022, ApJS, 259, 60, doi: 10.3847/1538-4365/ac5323

    2022 doi

  27. [35]

    C., & Casey, A

    Schlaufman, K. C., & Casey, A. R. 2014, ApJ, 797, 13, doi: 10.1088/0004-637X/797/1/13

    2014 doi

  28. [36]

    J., Finkbeiner, D

    Schlegel, D. J., Finkbeiner, D. P., & Davis, M. 1998, ApJ, 500, 525, doi: 10.1086/305772

    1998 doi

  29. [37]

    2016, A&A, 591, A118, doi: 10.1051/0004-6361/201628497

    Soubiran, C., Le Campion, J.-F., Brouillet, N., & Chemin, L. 2016, A&A, 591, A118, doi: 10.1051/0004-6361/201628497

    2016 doi

  30. [38]

    2017, MNRAS, 471, 2587, doi: 10.1093/mnras/stx1068 Str¨omgren, B

    Starkenburg, E., Martin, N., Youakim, K., et al. 2017, MNRAS, 471, 2587, doi: 10.1093/mnras/stx1068 Str¨omgren, B. 1964, Astrophysica Norvegica, 9, 333

    2017 doi

  31. [39]

    2022, Research in Astronomy and Astrophysics, 22, 105004, doi: 10.1088/1674-4527/ac8b5b

    Tan, K.-F., Zhao, G., Fan, Z., et al. 2022, Research in Astronomy and Astrophysics, 22, 105004, doi: 10.1088/1674-4527/ac8b5b

    2022 doi

  32. [40]

    2014, in Setting the scene for Gaia and LAMOST, ed

    Wang, W., Zhao, G., Chen, Y ., & Liu, Y . 2014, in Setting the scene for Gaia and LAMOST, ed. S. Feltzing, G. Zhao, N. A. Walton, & P. Whitelock, V ol. 298, 326–330, doi:10.1017/S1743921313006534

    2014 doi

  33. [41]

    D., Placco, V

    Whitten, D. D., Placco, V . M., Beers, T. C., et al. 2021, ApJ, 912, 147, doi: 10.3847/1538-4357/abee7e

    2021 doi

  34. [42]

    2022, ApJS, 263, 29, doi: 10.3847/1538-4365/ac9908

    Xu, S., Yuan, H., Zhang, R., et al. 2022, ApJS, 263, 29, doi: 10.3847/1538-4365/ac9908

    2022 doi

  35. [43]

    2022, A&A, 659, A181, doi: 10.1051/0004-6361/202142724

    Yang, L., Yuan, H., Xiang, M., et al. 2022, A&A, 659, A181, doi: 10.1051/0004-6361/202142724

    2022 doi

  36. [44]

    J., et al

    Yanny, B., Rockosi, C., Newberg, H. J., et al. 2009, AJ, 137, 4377, doi: 10.1088/0004-6256/137/5/4377

    2009 doi

  37. [45]

    E., Bessell, M

    Yong, D., Norris, J. E., Bessell, M. S., et al. 2013, ApJ, 762, 26, doi: 10.1088/0004-637X/762/1/26

    2013 doi

  38. [46]

    C., Placco, V

    Yoon, J., Beers, T. C., Placco, V . M., et al. 2016, ApJ, 833, 20, doi: 10.3847/0004-637X/833/1/20

    2016 doi

  39. [47]

    G., Adelman, J., Anderson, John E., J., et al

    York, D. G., Adelman, J., Anderson, John E., J., et al. 2000, AJ, 120, 1579, doi: 10.1086/301513

    2000 doi

  40. [48]

    2015, ApJ, 803, 13, doi: 10.1088/0004-637X/803/1/13

    Yuan, H., Liu, X., Xiang, M., Huang, Y ., & Chen, B. 2015, ApJ, 803, 13, doi: 10.1088/0004-637X/803/1/13

    2015 doi

  41. [49]

    2023, ApJS, 264, 14, doi: 10.3847/1538-4365/ac9dfa

    Zhang, R., & Yuan, H. 2023, ApJS, 264, 14, doi: 10.3847/1538-4365/ac9dfa

    2023 doi

  42. [50]

    2021, Research in Astronomy and Astrophysics, 21, 319, doi: 10.1088/1674-4527/21/12/319

    Zhang, R.-Y ., Yuan, H.-B., Liu, X.-W., et al. 2021, Research in Astronomy and Astrophysics, 21, 319, doi: 10.1088/1674-4527/21/12/319

    2021 doi

  43. [51]

    2012, Research in Astronomy and Astrophysics, 12, 723, doi: 10.1088/1674-4527/12/7/002

    Zhao, G., Zhao, Y .-H., Chu, Y .-Q., Jing, Y .-P., & Deng, L.-C. 2012, Research in Astronomy and Astrophysics, 12, 723, doi: 10.1088/1674-4527/12/7/002

    2012 doi

  44. [52]

    2018, Research in Astronomy and Astrophysics, 18, 147, doi: 10.1088/1674-4527/18/12/147 —

    Zheng, J., Zhao, G., Wang, W., et al. 2018, Research in Astronomy and Astrophysics, 18, 147, doi: 10.1088/1674-4527/18/12/147 —. 2019, Research in Astronomy and Astrophysics, 19, 003, doi: 10.1088/1674-4527/19/1/3

    2018 doi

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.