The RAdial Velocity Experiment (RAVE): Parameterisation of RAVE spectra based on convolutional neural networks

Abstract

In the context of large spectroscopic surveys of stars, data-driven methods are key in deducing physical parameters for millions of spectra in a short time. Convolutional neural networks (CNNs) enable us to connect observables (e.g. spectra, stellar magnitudes) to physical properties (atmospheric parameters, chemical abundances, or labels in general). We trained a CNN, adopting stellar atmospheric parameters and chemical abundances from APOGEE DR16 (resolution R=22500) data as training set labels. As input, we used parts of the intermediate-resolution RAVE DR6 spectra (R~7500) overlapping with the APOGEE DR16 data as well as broad-band ALL_WISE and 2MASS photometry, together with Gaia DR2 photometry and parallaxes. We derived precise atmospheric parameters Teff, log(g), and [M/H] along with the chemical abundances of [Fe/H], [alpha/M], [Mg/Fe], [Si/Fe], [Al/Fe], and [Ni/Fe] for 420165 RAVE spectra. The precision typically amounts to 60K in Teff, 0.06 in log(g) and 0.02-0.04 dex for individual chemical abundances. Incorporating photometry and astrometry as additional constraints substantially improves the results in terms of the accuracy and precision of the derived labels. We provide a catalogue of CNN-trained atmospheric parameters and abundances along with their uncertainties for 420165 stars in the RAVE survey. CNN-based methods provide a powerful way to combine spectroscopic, photometric, and astrometric data without the need to apply any priors in the form of stellar evolutionary models. The developed procedure can extend the scientific output of RAVE spectra beyond DR6 to ongoing and planned surveys such as Gaia RVS, 4MOST, and WEAVE. We call on the community to place a particular collective emphasis and on efforts to create unbiased training samples for such future spectroscopic surveys.

arXiv comment31 pages, 30 figures, accepted for publication in A&A, in Press

References in corpus (76)

• Gaia Data Release 2. Summary of the contents and survey properties
• The Wide-field Infrared Survey Explorer (WISE): Mission Description and Initial On-orbit Performance
• Gaia Data Release 2: The astrometric solution
• The Geneva-Copenhagen survey of the Solar neighbourhood: Ages, metallicities, and kinematic properties of 14,000 F and G dwarfs
• The Sixteenth Data Release of the Sloan Digital Sky Surveys: First Release from the APOGEE-2 Southern Survey and Full Release of eBOSS Spectra
• Astrophysics Source Code Library
• SEGUE: A Spectroscopic Survey of 240,000 stars with g=14-20
• The New Galaxy: Signatures of its Formation
• Exploring the Milky Way stellar disk. A detailed elemental abundance study of 714 F and G dwarf stars in the Solar neighbourhood
• The Radial Velocity Experiment (RAVE): first data release
• Elemental Abundance Survey of The Galactic Thick Disk
• The Chemical Compositions of Galactic Disk F and G Dwarfs
• ASPCAP: The Apogee Stellar Parameter and Chemical Abundances Pipeline
• An absolutely calibrated effective temperature scale from the InfraRed Flux Method
• Chemical Cartography with APOGEE: Metallicity Distribution Functions and the Chemical Structure of the Milky Way Disk
• Gaia Data Release 2: Catalogue validation
• Na-O Anticorrelation and HB. VIII. Proton-capture elements and metallicities in 17 globular clusters from UVES spectra
• Chemical abundances of 1111 FGK stars from the HARPS GTO planet search program.Galactic stellar populations and planets
• The Radial Velocity Experiment (RAVE): Fifth Data Release
• APOGEE Data and Spectral Analysis from SDSS Data Release 16: Seven Years of Observations Including First Results from APOGEE-South
• A Spectroscopic Study of the Ancient Milky Way: F- and G-Type Stars in the Third Data Release of the Sloan Digital Sky Survey
• Gaia Data Release 1: Catalogue validation
• The Cannon: A data-driven approach to stellar label determination
• The RAdial Velocity Experiment (RAVE): Fourth data release
• The GALAH Survey: Second Data Release
• Chemical Abundances for 855 Giants in the Globular Cluster Omega Centauri (NGC 5139)
• The Second APOKASC Catalog: The Empirical Approach
• The Radial Velocity Experiment (RAVE): second data release
• Photo-astrometric distances, extinctions, and astrophysical parameters for Gaia DR2 stars brighter than G = 18
• The Pristine survey I: Mining the Galaxy for the most metal-poor stars
• StarHorse: A Bayesian tool for determining stellar masses, ages, distances, and extinctions for field stars
• The Payne: self-consistent ab initio fitting of stellar spectra
• High-Performance Neural Networks for Visual Object Classification
• Formation and Evolution of the Disk System of the Milky Way: [alpha/Fe] Ratios and Kinematics of the SEGUE G-Dwarf Sample
• Deep learning of multi-element abundances from high-resolution spectroscopic data
• Accurate fundamental parameters for Lower Main Sequence Stars
• Accuracy and precision of industrial stellar abundances
• From the bulge to the outer disc: StarHorse stellar parameters, distances, and extinctions for stars in APOGEE DR16 and other spectroscopic surveys
• The Sixth Data Release of the Radial Velocity Experiment (RAVE) -- II: Stellar Atmospheric Parameters, Chemical Abundances and Distances
• Abundance trends in kinematical groups of the Milky Way's disk
• The Sixth Data Release of the Radial Velocity Experiment (RAVE) -- I: Survey Description, Spectra and Radial Velocities
• Automated derivation of stellar atmospheric parameters and chemical abundances: the MATISSE algorithm
• Abundances for metal-poor stars with accurate parallaxes II. alpha-elements in the halo
• An Application of Deep Neural Networks in the Analysis of Stellar Spectra
• The GALAH survey: An abundance, age, and kinematic inventory of the solar neighbourhood made with TGAS
• A spectroscopic survey of thick disc stars outside the solar neighbourhood
• Spectro-photometric distances to stars: a general-purpose Bayesian approach
• Detailed chemical composition of the open cluster IC 4651: The iron peak, alpha elements, and Li
• A new stellar chemo-kinematic relation reveals the merger history of the Milky Way disc
• The Rich Are Different: Evidence from the RAVE Survey for Stellar Radial Migration
• That's How We Roll: The NASA K2 Mission Science Products and Their Performance Metrics
• Automatic stellar spectra parameterisation in the IR CaII triplet region
• Metal-Poor Lithium-Rich Giants in the Radial Velocity Experiment Survey
• Chemical gradients in the Milky Way from the RAVE data. I. Dwarf stars
• The GALAH survey: The data reduction pipeline
• The RAVE-on catalog of stellar atmospheric parameters and chemical abundances for chemo-dynamic studies in the Gaia era
• The Gaia-ESO Survey: Sodium and aluminium abundances in giants and dwarfs - Implications for stellar and Galactic chemical evolution
• The RAVE catalogue of stellar elemental abundances: first data release
• The AMBRE Project: Constraining the lithium evolution in the Milky Way
• Chemical gradients in the Milky Way from the RAVE data. II. Giant stars
• Fundamental stellar parameters and metallicities from Bayesian spectroscopy. Application to low- and high-resolution spectra
• The Pristine survey III: Spectroscopic confirmation of an efficient search for extremely metal-poor stars
• The relation between chemical abundances and kinematics of the Galactic disc with RAVE
• Origins of the Thick Disk as Traced by the Alpha-Elements of Metal-Poor Giant Stars Selected from RAVE
• The Cannon 2: A data-driven model of stellar spectra for detailed chemical abundance analyses
• Improved distances and ages for stars common to TGAS and RAVE
• Elemental abundances in the Blanco 1 open cluster
• Very metal-poor stars observed by the RAVE survey
• Assessing the accuracy of the gravity determination in late-type stars with solar-like pulsators
• RAVE stars in K2 - I. Improving RAVE red giants spectroscopy using asteroseismology from K2 Campaign 1
• Yule-Simpson's paradox in Galactic Archaeology
• ATHOS: On-the-fly stellar parameter determination of FGK stars based on flux ratios from optical spectra
• Assessing the performance of LTE and NLTE synthetic stellar spectra in a machine learning framework
• Calibration of LAMOST Stellar Surface Gravities Using the Kepler Asteroseismic Data
• Correlations between age, kinematics, and chemistry as seen by the RAVE survey
• Asymmetric metallicity patterns in the stellar velocity space with RAVE

Cited by in corpus (17)

• Assessment of [Fe/H] determinations for FGK stars in spectroscopic surveys
• Recovery of TESS Stellar Rotation Periods Using Deep Learning
• The Spectroscopic Binaries from LAMOST Medium-Resolution Survey (MRS). I. Searching for Double-lined Spectroscopic Binaries (SB2s) with Convolutional Neural Network
• APOGEE Net: An expanded spectral model of both low mass and high mass stars
• The distribution of [$\alpha$/Fe] in the Milky Way disc
• Cycle-StarNet: Bridging the gap between theory and data by leveraging large datasets
• Transferring spectroscopic stellar labels to 217 million Gaia DR3 XP stars with SHBoost
• Titans metal-poor reference stars. I. Accurate effective temperatures and surface gravities for dwarfs and subgiants from 3D non-LTE H$\alpha$ profiles and Gaia parallaxes
• Stellar Parameters and Chemical Abundances Estimated from LAMOST-II DR8 MRS based on Cycle-StarNet
• The Gaia-ESO Survey: Chemical evolution of Mg and Al in the Milky Way with Machine-Learning
• The Gaia-ESO Survey: Preparing the ground for 4MOST & WEAVE galactic surveys. Chemical evolution of lithium with machine learning
• The Swan: Data-Driven Inference of Stellar Surface Gravities for Cool Stars from Photometric Light Curves
• Inferring stellar parameters and their uncertainties from high-resolution spectroscopy using invertible neural networks
• StarUnLink: identifying and mitigating signals from communications satellites in stellar spectral surveys
• Playing CHESS with stars. I. Search for similar stars in large spectroscopic data sets
• A method to derive self-consistent NLTE astrophysical parameters for 4 million high-resolution 4MOST stellar spectra in half a day with invertible neural networks
• Exploring Late Stellar Evolution in the Era of Large Surveys: Machine Learning Prospects for Hot Subdwarfs and White Dwarfs