| 1. |
- Hourihane, A., et al.
(författare)
-
The Gaia-ESO Survey : Homogenisation of stellar parameters and elemental abundances
- 2023
-
Ingår i: Astronomy and Astrophysics. - : EDP Sciences. - 0004-6361 .- 1432-0746. ; 676
-
Tidskriftsartikel (refereegranskat)abstract
- The Gaia-ESO Survey is a public spectroscopic survey that targeted greater than or similar to 10(5) stars covering all major components of the Milky Way from the end of 2011 to 2018, delivering its final public release in May 2022. Unlike other spectroscopic surveys, Gaia-ESO is the only survey that observed stars across all spectral types with dedicated, specialised analyses: from O (T-eff similar to 30 000-52 000 K) all the way to K-M (greater than or similar to 3500 K). The physics throughout these stellar regimes varies significantly, which has previously prohibited any detailed comparisons between stars of significantly different types. In the final data release (internal data release 6) of the Gaia-ESO Survey, we provide the final database containing a large number of products, such as radial velocities, stellar parameters and elemental abundances, rotational velocity, and also, for example, activity and accretion indicators in young stars and membership probability in star clusters for more than 114 000 stars. The spectral analysis is coordinated by a number of working groups (WGs) within the survey, each specialised in one or more of the various stellar samples. Common targets are analysed across WGs to allow for comparisons (and calibrations) amongst instrumental setups and spectral types. Here we describe the procedures employed to ensure all survey results are placed on a common scale in order to arrive at a single set of recommended results for use by all survey collaborators. We also present some general quality and consistency checks performed on the entirety of the survey results.
|
|
| 2. |
- Ambrosch, M., et al.
(författare)
-
The Gaia -ESO Survey : Chemical evolution of Mg and Al in the Milky Way with machine learning
- 2023
-
Ingår i: Astronomy and Astrophysics. - : EDP Sciences. - 0004-6361 .- 1432-0746. ; 672
-
Tidskriftsartikel (refereegranskat)abstract
- Context. To take full advantage of upcoming large-scale spectroscopic surveys, it will be necessary to parameterize millions of stellar spectra in an efficient way. Machine learning methods, especially convolutional neural networks (CNNs), will be among the main tools geared at achieving this task. Aims. We aim to prepare the groundwork for machine learning techniques for the next generation of spectroscopic surveys, such as 4MOST and WEAVE. Our goal is to show that CNNs can predict accurate stellar labels from relevant spectral features in a physically meaningful way. The predicted labels can be used to investigate properties of the Milky Way galaxy. Methods. We built a neural network and trained it on GIRAFFE spectra with their associated stellar labels from the sixth internal Gaia-ESO data release. Our network architecture contains several convolutional layers that allow the network to identify absorption features in the input spectra. The internal uncertainty was estimated from multiple network models. We used the t-distributed stochastic neighbor embedding tool to remove bad spectra from our training sample. Results. Our neural network is able to predict the atmospheric parameters Teff and log(g) as well as the chemical abundances [Mg/Fe], [Al/Fe], and [Fe/H] for 36 904 stellar spectra. The training precision is 37 K for Teff, 0.06 dex for log(g), 0.05 dex for [Mg/Fe], 0.08 dex for [Al/Fe], and 0.04 dex for [Fe/H]. Network gradients reveal that the network is inferring the labels in a physically meaningful way from spectral features. We validated our methodology using benchmark stars and recovered the properties of different stellar populations in the Milky Way galaxy. Conclusions. Such a study provides very good insights into the application of machine learning for the analysis of large-scale spectroscopic surveys, such as WEAVE and 4MOST Milky Way disk and bulge low- and high-resolution (4MIDABLE-LR and -HR). The community will have to put substantial efforts into building proactive training sets for machine learning methods to minimize any possible systematics.
|
|
