Abundances in eight bulge stars from the optical and near-infrared

Abundances in the Galactic bulge are being scrutinized for a better definition of the fast star formation rate and identification of the supernovae that enriched the medium very early on (e.g., Barbuy et al. 2018a; Matteucci 2021). This applies to field and globular cluster stars, with the latter also involving new progress in our understanding of self-enrichment and the multiple stellar populations that undoubtedly occurred in almost all clusters (Piotto et al., 2015; Renzini et al., 2015; Milone et al., 2017; Bastian & Lardo, 2018; Milone & Marino, 2022).

The light elements are discriminators of the type of supernova and nucleosynthesis that formed them, and in order to use them, reliable abundance values are needed. Consequently, in this work, we investigate the possible differences between the abundance derivation from the optical and the near-infrared (NIR) for eight stars analyzed in the optical by our group and also available from NIR spectra.

Instrumentation in astronomy allowing observations in the NIR has progressed significantly in the recent past. Apache Point Observatory Galactic Evolution Experiment (APOGEE– Majewski et al. 2017) allows high-resolution spectroscopy in the H-band at a resolution of R $\sim$ 22 000 carried out at the 2.5m Sloan Foundation Telescope at the Apache Point Observatory in New Mexico (APOGEE-1 and APOGEE-2N: Beaton et al. 2021), and the 2.5m du Pont Telescope at Las Campanas Observatory in Chile (APOGEE-2S). Spectroscopy in the JHK bands will be intensified with the James Webb Space Telescope (JWST), and new spectrographs on ground-based telescopes such as the Multi-Object Optical and Near-IR Spectrograph (MOONS) at the Very Large Telescope (VLT), the Warm INfrared Echelle spectrograph to REalize Decent high-resolution spectroscopy (WINERED) at the Magellan telescope (see Minniti et al. (2024)), and the Multi-Object Spectrograph for Astrophysics, Intergalactic-medium studies and Cosmology (MOSAIC) at the Extremely Large Telescope (ELT) at the European Southern Observatory (ESO), among others.

The aim of the present work is to compare available lines and results that can be obtained in the optical and H-band for a list of elements. This study is useful as it indicates the most reliable lines in the optical and/or H-band —which is of particular interest for observations with the ELT/MOSAIC that should allow simultaneous observations in the NIR and selected regions in the optical at resolutions of 6,000 and 20,000 (Hammer et al., 2021).

The analysis of H-band spectra in the APOGEE project is carried out through a Nelder-Mead algorithm (Nelder & Mead, 1965), which simultaneously fits the stellar parameters effective temperature (T_eff, gravity (log g), metallicity ([Fe/H]), and microturbulence velocity (v_t), together with the abundances of carbon, nitrogen, and alpha-elements with the APOGEE Stellar Parameter and Chemical Abundances Pipeline - ASPCAP pipeline, which is based on the FERRE code (García-Pérez et al., 2016). Spectroscopists within the APOGEE community are carrying out continuous and intensive work, but improvements are still needed for some elements and lines.

Jönsson et al. (2018) carried out an important verification of the accuracy of the ASPCAP results, and reported previous analyses with the same intent: Hawkins et al. (2016) used the DR12 APOGEE to analyze stars with asteroseismic analysis from Kepler light curves, and found differences regarding Si, S, Ti, and V. Souto et al. (2016) used the DR13 APOGEE to analyze spectra of 12 giant stars belonging to the open cluster NGC 2420 ([Fe/H] $\sim$ -0.16), and found differences in the abundances of Na, Al, and V. Fernández-Trincado et al. (2016) manually reanalyzed the DR12 APOGEE spectrum of one peculiar metal-poor field giant star with a globular cluster second-generation abundance pattern, and found differences in the abundances of C, N, O, Mg, and Al of about 0.3 dex.

Holtzman et al. (2018) described the series of technical steps of the ASPCAP procedure, and the grids of synthetic spectra used to derive stellar parameters. Jönsson et al. (2018) carried out a comparison between results from DR13 and DR14, as well as selected literature results, with mean differences reported in their Table 5. Jönsson et al. (2020) and Hayes et al. (2022) studied the details of several lines from DR16 and DR17 results, and pointed out problems and limitations.

In Razera et al. (2022), our group analyzed 58 moderately metal-poor bulge field stars ($-$2.0 $<$ [Fe/H] $<$ $-$0.8) from APOGEE spectra, and identified the most suitable spectral lines for deriving the abundances of C, N, O, Mg, Si, Ca, and Ce given that some of the ASPCAP adopted lines are too faint in these metal-poor stars. These authors confirmed the Mg, Si, and Ca from ASPCAP, whereas O and Ce were found to be more abundant than the ASPCAP values.

In the present work, we identified eight stars previously analyzed by our group from optical high-resolution spectra collected with the Ultraviolet and Visual Echelle Spectrograph (UVES) at the VLT. Importantly, these sources were also observed with APOGEE. For this sample, we carried out a detailed comparison of stellar parameters and abundance results and inspected the available lines for a list of elements.

In Section 2, we list the sample stars and report previous calculations. In Section 3, we describe the present calculations. In Section 4 we compare the stellar parameters and abundances derived from the optical and near-infrared, and show how we carried out fits to lines both in the optical and in the near-infrared. We discuss our results in Section 5 and draw conclusions in Section 6.

The sample consists of eight stars analyzed in the optical using the UVES spectrograph at the VLT (ESO), and for which there are available observations with APOGEE. These stars are from the following globular clusters: one star from HP 1 (Barbuy et al., 2016), two stars from NGC 6558 (Barbuy et al., 2018a), one star from NGC 6522 (Barbuy et al., 2014), and one star from Palomar 6 (Souza et al., 2021). In Ernandes et al. (2018) the abundances of Sc, V, Mn, Cu, and Zn were derived for most of these stars. The sample is completed with three bulge field stars analyzed for different elements, namely C, N, and O in Zoccali et al. (2006); Lecureur et al. (2007); Jönsson et al. (2017), and Friaça & Barbuy (2017) (the latter is adopted), Mn and Zn in Barbuy et al. (2013, 2015) and da Silveira et al. (2018), and Co and Cu in Ernandes et al. (2020).

The H-band data are from the APOGEE project, which is part of the Sloan Digital Sky Survey III and IV (SDSS; Blanton et al. 2017). The project aimed to investigate Milky Way stars observed at high resolution and high signal-to-noise ratios (S/Ns) in the H-band. The project SDSS-IV technical summary, the SDSS telescope, and APOGEE spectrograph are described in Blanton et al. (2017); Gunn et al. (2006); Wilson et al. (2019), respectively, whereas Zasowski et al. (2013, 2017); Beaton et al. (2021); Santana et al. (2021) describe the APOGEE and APOGEE-2 target selections. The data release 17 (DR17) contains high-resolution (R $\sim$ 22,500 ) H-band spectra (15140-16940 Å) for about 7$\times$10⁵ stars. The 2.5m Sloan Foundation Telescope in New Mexico, USA, and the 2.5m Irenee du Pont Telescope in Las Campanas Observatory in Chile were used, covering the northern and southern skies respectively. In the present work, we reanalyze the optical spectra for the elements Na, Mg, Al, Si, Ca, and Ti, and in the H-band the elements C, N, O, Mg, Al, Si, and Ca for the lines reported in Table 1.

We recomputed the abundances of C, N, O, Mg, Al, Si, and Ca in the H-band, and Na, Mg, Al, Si, Ca, and Ti in the optical using the code TURBOSPECTRUM from Alvarez & Plez (1998) and Plez (2012). Model atmosphere grids are from Gustafsson et al. (2008). The solar abundances are from Grevesse & Sauval (1998): A(C)=8.55, A(N)=7.97, from Steffen et al. (2015): A(O)=8.76, and from Smith et al. (2021): A(Na)=6.17, A(Mg)=7.53, A(Al)=6.37, A(Si)=7.51, A(Ca)=6.31, A(Ti)=4.90.

Table 1 reports the lines in the H-band that we verified in the spectra of the eight sample stars. The atomic line list employed is that from the APOGEE collaboration, together with the molecular lines described in Smith et al. (2021). The lines of the studied elements C, N, O, Na, Al, Mg, Si, and Ca (Ti is excluded) were recently examined by Razera et al. (2022) and Barbuy et al. (2023).

In the optical, the lines measured are listed, for example, in Barbuy et al. (2016). For the calculations, we adopted the line list from VALD3 (Ryabchikova et al., 2015). The new VALD3 line list includes hyperfine structure when possible and needed. The molecular lines include CH from Masseron et al. (2014), CN from Brooke et al. (2014), C2 from Brooke et al. (2013), MgH from Skory et al. (2003), and TiO, including the $\alpha$, $\beta$, $\gamma$, $\gamma$’, $\epsilon$, $\delta$ (and $\phi$ in the NIR), from Jorgensen (1994).

Below we give a brief description of the methods employed in the analysis of optical and H-band spectra in order to make clear that there are differences between them due to the different lines available in each spectral region.

In the optical, initial parameters based on photometric indicators are adopted, and iteratively modified based on ionization and excitation equilibra of Fe I and Fe II lines of different ionization and excitation energies until the parameters converge. The excitation equilibrium requires a constant Fe abundance for Fe I lines with different excitation energies $\chi_{\rm exc}$, and is affected mainly by iterating the value of the effective temperature. The ionization equilibrium requires similar Fe abundances from Fe I and Fe II lines, which are affected mainly by the gravity log g value. Finally, the microturbulence velocity v_t is obtained by imposing a constant Fe abundance versus equivalent width (EW) of Fe I lines.

In the H-band, the Fe I lines have almost all of their excitation energies in the range 5.5 to 7.0 eV, which means that the concept of excitation equilibrium cannot be used as initial temperature indicator. Therefore, initial indication for the effective temperature are photometric indices. The ASPCAP pipeline derives their noncalibrated stellar parameters, in this case from Data Release 17 (DR17), using a spectroscopic solution that minimizes the chi-squared differences between the observations and the models in the following seven dimensions: T_eff, log g, [Fe/H], v_t, [$\alpha$/Fe], [C/Fe], and [N/Fe]). As we demonstrate below, the fit to molecular lines of CN, CH, CO, and OH is a powerful tool for constraining stellar parameters, in particular the effective temperature.

In the following sections, we carry out abundance derivations using the two sets of stellar parameters —that is, those obtained with the APOGEE ASPCAP procedure and the detailed analyses with the optical spectra observed with UVES— that are reported in Table 3.

In this section, we compare effective temperature T_eff, gravity log g, metallicity [Fe/H], and microturbulence velocity v_t obtained from analyses of spectra in the optical and the H-band, and element abundances derived in both cases.

In this work, we fitted the lines in the optical and H-band and compared the results between these two sets. If the resulting abundances of some lines needed for the fits are too overabundant or too deficient, this indicates that the stellar parameters are not suitable.

In this work, we compared abundance results obtained from optical spectra (VLT/UVES) and H-band spectra (APOGEE) for eight bulge stars. The calculations were carried out adopting stellar parameters from the literature obtained from optical and H-band spectra. We then computed abundances employing the TURBOSPECTRUM code throughout for the sake of homogeneity, adopting the two sets of stellar parameters for the lines in the optical and the H-band. The aim of this work is to find the best methods for deriving stellar parameters and their advantages and disadvantages, and to identify the best lines for each of the analyzed elements.

A first and important conclusion is that the stellar parameters —and in particular the effective temperatures obtained from ASPCAP— that correspond to a spectroscopic solution minimizing the chi-squared differences between the observations and the models in seven dimensions, lead to reliable parameters. This is due to the derivation of C, N, and O abundances through the strength of molecular lines of CO, CN, and OH, leading to reliable effective temperature, and more so when dealing with high-S/N spectra. This is an important compensation for the lack of Fe I lines of varied excitation energies, which are used to constrain effective temperatures in the optical.

In conclusion, as concerns stellar parameters, previous APOGEE and UVES results are suitable for stars HP1-2, NGC6522-402361 and Palomar6-401. A toohigh effective temperature was estimated for the stars N6558-283 and NGC6558-1160 from the optical, therefore the ASPCAP parameters are more suitable. For the 3 metal-rich stars the optical parameters derived are more suitable than the ASPCAP ones.

Regarding the atomic lines, the Mg, Al, and Si lines in both H-band and optical are suitable, with Si having stronger lines in the H-band than in the optical. For Ca, the optical appears to have a greater number of reliable lines. Na and Ti do not have reliable lines in the H-band, and therefore the optical lines are more suitable.

This findings of study will be particularly useful for future observations with the ELT/MOSAIC spectrograph, which will allow simultaneous observations of the H-band and selected regions in the optical.

Finally, a comparison of abundances in the presently studied 8 stars with bulge field stars by Queiroz et al. (2021) shows good agreement,and despite the limited number of stars in our sample, they cover the entire bulge metallicity range.

PS acknowledges Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) post-doctoral fellowships 2020/13239-5 and 2022/14382-1. BB acknowledges grants from FAPESP, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) - Financial code 001. SOS acknowledges a FAPESP PhD fellowship no. 2018/22044-3. PS, BB, HE, and SOS are part of the Brazilian Participation Group (BPG) in the Sloan Digital Sky Survey (SDSS), from the Laboratório Interinstitucional de e-Astronomia – LIneA, Brazil. J.G.F-T gratefully acknowledges the grant support provided by Proyecto Fondecyt Iniciación No. 11220340, Proyecto Fondecyt Postdoc No. 3230001 (Sponsored by J.G.F-T) and from the Joint Committee ESO-Government of Chile 2021 (ORP 023/2021), and 2023 (ORP 062/2023). D.G-D gratefully acknowledges support from the Joint Committee ESO-Government of Chile 2021 (ORP 023/2021), as well as the support from call N° 785 of 2017 of the Colombian Departamento Administrativo de Ciencia, Tecnología e Innovación (COLCIENCIAS). Apogee project: Funding for the Sloan Digital Sky Survey IV has been provided by the Alfred P. Sloan Foundation, the U.S. Department of Energy Office of Science, and the Participating Institutions. SDSS acknowledges support and resources from the Center for High-Performance Computing at the University of Utah. The SDSS web site is www.sdss.org. SDSS is managed by the Astrophysical Research Consortium for the Participating Institutions of the SDSS Collaboration including the Brazilian Participation Group, the Carnegie Institution for Science, Carnegie Mellon University, Center for Astrophysics — Harvard & Smithsonian (CfA), the Chilean Participation Group, the French Participation Group, Instituto de Astrofísica de Canarias, The Johns Hopkins University, Kavli Institute for the Physics and Mathematics of the Universe (IPMU) / University of Tokyo, the Korean Participation Group, Lawrence Berkeley National Laboratory, Leibniz Institut für Astrophysik Potsdam (AIP), Max-Planck-Institut für Astronomie (MPIA Heidelberg), Max-Planck-Institut für Astrophysik (MPA Garching), Max-Planck-Institut für Extraterrestrische Physik (MPE), National Astronomical Observatories of China, New Mexico State University, New York University, University of Notre Dame, Observatório Nacional / MCTI, The Ohio State University, Pennsylvania State University, Shanghai Astronomical Observatory, United Kingdom Participation Group, Universidad Nacional Autónoma de México, University of Arizona, University of Colorado Boulder, University of Oxford, University of Portsmouth, University of Utah, University of Virginia, University of Washington, University of Wisconsin, Vanderbilt University, and Yale University.

The spectra observed in the optical with the UVES spectrograph are available at the archival data from the European Southern Observatory (ESO). The spectra observed in the H-band are from the APOGEE survey.