Transferring spectroscopic stellar labels to 217 million <i>Gaia</i> DR3 XP stars with SHBoost

With Gaia Data Release 3 (DR3), new and improved astrometric, photometric, and spectroscopic measurements for 1.8 billion stars have become available. Alongside this wealth of new data, however, there are challenges in finding efficient and accurate computational methods for their analysis. In this paper, we explore the feasibility of using machine learning regression as a method of extracting basic stellar parameters and line-of-sight extinctions from spectro-photometric data. To this end, we built a stable gradient-boosted random-forest regressor ( xgboost ), trained on spectroscopic data, capable of producing output parameters with reliable uncertainties from Gaia DR3 data (most notably the low-resolution XP spectra), without ground-based spectroscopic observations. Using Shapley additive explanations, we interpret how the predictions for each star are influenced by each data feature. For the training and testing of the network, we used high-quality parameters obtained from the StarHorse code for a sample of around eight million stars observed by major spectroscopic stellar surveys, complemented by curated samples of hot stars, very metal-poor stars, white dwarfs, and hot sub-dwarfs. The training data cover the whole sky, all Galactic components, and almost the full magnitude range of the Gaia DR3 XP sample of more than 217 million objects that also have reported parallaxes. We have achieved median uncertainties of 0.20 mag in V-band extinction, 0.01 dex in logarithmic effective temperature, 0.20 dex in surface gravity, 0.18 dex in metallicity, and 12% in mass (over the full Gaia DR3 XP sample, with considerable variations in precision as a function of magnitude and stellar type). We succeeded in predicting competitive results based on Gaia DR3 XP spectra compared to classical isochrone or spectral-energy distribution fitting methods we employed in earlier works, especially for parameters A V and T eff , along with the metallicity values. Finally, we showcase some potential applications of this new catalogue, including extinction maps, metallicity trends in the Milky Way, and extended maps of young massive stars, metal-poor stars, and metal-rich stars.

Authors

Institutions

Publication Details

Journal
Astronomy and Astrophysics
Published
2024-09-30
DOI
https://doi.org/10.1051/0004-6361/202451427
Citations
24
Primary Topic
Astronomy and Astrophysical Research
Type
article
Field-Weighted Citation Impact
12.15

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Transferring spectroscopic stellar labels to 217 million Gaia DR3 XP stars with SHBoost

H. Enke, S. Malhotra, S. Nepal, M. Valentini et al.
24 citations
Astronomy and Astrophysics
Astronomy and Astrophysical Research
12.15
article

Transferring spectroscopic stellar labels to 217 million Gaia DR3 XP stars with SHBoost

H. Enke, S. Malhotra, S. Nepal, M. Valentini, C. Jordi, A. B. A. Queiroz, L. Casamiquela, A. Khalatyan, Ó. Jiménez-Arranz, F. Anders, C. Chiappini, G. Guiglion, M. Dal Ponte, J. Ardèvol, Matthias Steinmetz, M. Pantaleoni-González, G. Torralba Elipe
article en
24 citations

Abstract

With Gaia Data Release 3 (DR3), new and improved astrometric, photometric, and spectroscopic measurements for 1.8 billion stars have become available. Alongside this wealth of new data, however, there are challenges in finding efficient and accurate computational methods for their analysis. In this paper, we explore the feasibility of using machine learning regression as a method of extracting basic stellar parameters and line-of-sight extinctions from spectro-photometric data. To this end, we built a stable gradient-boosted random-forest regressor ( xgboost ), trained on spectroscopic data, capable of producing output parameters with reliable uncertainties from Gaia DR3 data (most notably the low-resolution XP spectra), without ground-based spectroscopic observations. Using Shapley additive explanations, we interpret how the predictions for each star are influenced by each data feature. For the training and testing of the network, we used high-quality parameters obtained from the StarHorse code for a sample of around eight million stars observed by major spectroscopic stellar surveys, complemented by curated samples of hot stars, very metal-poor stars, white dwarfs, and hot sub-dwarfs. The training data cover the whole sky, all Galactic components, and almost the full magnitude range of the Gaia DR3 XP sample of more than 217 million objects that also have reported parallaxes. We have achieved median uncertainties of 0.20 mag in V-band extinction, 0.01 dex in logarithmic effective temperature, 0.20 dex in surface gravity, 0.18 dex in metallicity, and 12% in mass (over the full Gaia DR3 XP sample, with considerable variations in precision as a function of magnitude and stellar type). We succeeded in predicting competitive results based on Gaia DR3 XP spectra compared to classical isochrone or spectral-energy distribution fitting methods we employed in earlier works, especially for parameters A V and T eff , along with the metallicity values. Finally, we showcase some potential applications of this new catalogue, including extinction maps, metallicity trends in the Milky Way, and extended maps of young massive stars, metal-poor stars, and metal-rich stars.

Astronomy and AstrophysicsVol. 691
Universidade da Coruña (ES), Centre National de la Recherche Scientifique (FR), University of Vienna (AT), Universidad de La Laguna (ES), University of Potsdam (DE), Lund University (SE), Heidelberg University (DE), Université Paris Sciences et Lettres (FR), Leibniz Institute for Astrophysics Potsdam (DE), Observatoire de Paris (FR), Institut d'Estudis Espacials de Catalunya (ES), Galaxies, Etoiles, Physique et Instrumentation (FR), Max Planck Institute for Astronomy (DE), Universidad Internacional De La Rioja (ES), Centro de Astrobiología (ES), Instituto de Astrofísica de Canarias (ES), Osservatorio Astronomico di Padova (IT), Universitat Politècnica de València (ES), Universitat de Barcelona (ES), Universitat Politècnica de Catalunya (ES)
National Science Foundation, Smithsonian Institution, U.S. Department of Energy, National Aeronautics and Space Administration, Alfred P. Sloan Foundation, W. M. Keck Foundation, Yale University, Carnegie Institution of Washington, Vanderbilt University, Ohio State University, University of Utah, University of Washington, Johns Hopkins University, Carnegie Mellon University, University of Notre Dame, University of Portsmouth, New Mexico State University, European Space Agency, York University, University of Oxford, European Commission, Macquarie University, Deutsche Forschungsgemeinschaft, Agence Nationale de la Recherche, Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, Chinese Academy of Sciences, Ministério da Ciência, Tecnologia e Inovação, Conselho Nacional de Desenvolvimento Científico e Tecnológico, Javna Agencija za Raziskovalno Dejavnost RS, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, Financiadora de Estudos e Projetos, Ministerio de Ciencia e Innovación, Istituto Nazionale di Astrofisica, Universidad Nacional Autónoma de México, Universitat de Barcelona, National Development and Reform Commission, Leibniz-Institut für Astrophysik Potsdam, Max-Planck-Institut für Astrophysik, Max-Planck-Institut für Astronomie, Instituto de Astrofísica de Canarias, Office of Science, University of Colorado Boulder, Smithsonian Astrophysical Observatory, Natural Sciences and Engineering Research Council of Canada, European Research Council, Leibniz-Gemeinschaft, Australian Astronomical Optics-Macquarie, European Regional Development Fund, Agencia Estatal de Investigación, Institut de Ciències del Cosmos, National Astronomical Observatories, Chinese Academy of Sciences, Lawrence Berkeley National Laboratory
Openalex Percentile: Top 1%
Astronomy and Astrophysical Research
12.15
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.