Search

We identified a sample of 58 candidate stars with metallicity [Fe/H] ≲ −0.8 that likely belong to the old bulge spheroid stellar population, and analyse their Na and Al abundances from Apache Point Observatory Galactic Evolution Experiment (APOGEE) spectra. In a previous work, we inspected APOGEE-Stellar Parameter and Chemical Abundance Pipeline abundances of C, N, O, Mg, Al, Ca, Si, and Ce in this sample. Regarding Na lines, one of them appears very strong in about 20 per cent of the sample stars, but it is not confirmed by other Na lines, and can be explained by sky lines, which affect the reduced spectra of stars in a certain radial velocity range. The Na abundances for 15 more reliable cases were taken into account. Al lines in the H band instead appear to be very reliable. Na and Al exhibit a spread in abundances, whereas no spread in N abundances is found, and we found no correlation between them, indicating that these stars could not be identified as second-generation stars that originated in globular clusters. We carry out the study of the behaviour of Na and Al in our sample of bulge stars and literature data by comparing them with chemodynamical evolution model suitable for the Galactic bulge. The Na abundances show a large spread, and the chemodynamical models follow the main data, whereas for aluminum instead, the models reproduce very satisfactorily the nearly secondary-element behaviour of aluminum in the metallicity range below [Fe/H] ≲ −1.0. For the lower-metallicity end ([Fe/H < −2.5), hypernovae are assumed to be the main contributor to yields. [ABSTRACT FROM AUTHOR]

Context. Abundances of slow neutron-capture process (s-process) elements in stars with exquisite asteroseismic, spectroscopic, and astrometric constraints offer a novel opportunity to study stellar evolution, nucleosynthesis, and Galactic chemical evolution. Aims. We investigate one of the least studied s-process elements in the literature, cerium (Ce), using stars with asteroseismic constraints from the Kepler, K2, and TESS missions. Methods. We combined the global asteroseismic parameters derived from precise light curves obtained by the Kepler, K2, and TESS missions with stellar parameters and chemical abundances from the latest data release of the large spectroscopic survey APOGEE and astrometric data from the Gaia mission. Finally, we computed stellar ages using the code PARAM with a Bayesian estimation method. Results. We investigated the different trends of [Ce/Fe] as a function of metallicity, [α/Fe], and age taking into account the dependence on the radial position, especially in the case of K2 targets, which cover a wide galactocentric range. We finally explored the [Ce/α] ratios as a function of age in different galactocentric intervals. Conclusions The studied trends display a strong dependence of the Ce abundances on the metallicity and star formation history. The [Ce/Fe] ratio shows a non-monotonic dependence on [Fe/H] with a peak around −0.2 dex. Moreover, younger stars have higher [Ce/Fe] and [Ce/α] ratios than older stars, confirming the latest contribution of low- and intermediate-mass asymptotic giant branch stars to the Galactic chemical enrichment. In addition, the trends of [Ce/Fe] and [Ce/α] with age become steeper moving towards the outer regions of the Galactic disc, demonstrating more intense star formation in the inner regions than in the outer regions. Cerium is thus a potentially interesting element to help constrain stellar yields and the inside-out formation of the Milky Way disc. However, the large scatter in all the relations studied here suggests that spectroscopic uncertainties for this element are still too large. [ABSTRACT FROM AUTHOR]

Context. To take full advantage of upcoming large-scale spectroscopic surveys, it will be necessary to parameterize millions of stellar spectra in an effcient 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 Te 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 37K for Te, 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 MilkyWay 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. [ABSTRACT FROM AUTHOR]