Scientists are increasingly focused on refining abundance determination techniques for stars, and this research presents a crucial step forward in validating infrared spectral lines for this purpose. Scarlet Elgueta, Paula Jofré, and Claudia Aguilera-Gómez from the Instituto de Estudios Astrofísicos, Facultad de Ingeniería y Ciencias, Universidad Diego Portales, led a collaborative effort involving Ditte Slumstrup from Gran Telescopio CANARIAS, Álvaro Rojas-Arriagada from Departamento de Física, Universidad de Santiago de Chile, Ulrike Heiter from Observational Astrophysics, Department of Physics and Astronomy, Uppsala University, Laia Casamiquela working with colleagues at LIRA, Observatoire de Paris, Université PSL, Sorbonne Université, Université Paris Cité, CY Cergy Paris Université, CNRS, Manuela Zoccali from Instituto de Astrofísica, Pontificia Universidad Católica de Chile, Clare Worley from School of Physical and Chemical Sciences, Te Kura Mat u, University of Canterbury, and Caroline Soubiran from Laboratoire d’Astrophysique de Bordeaux, Univ. Bordeaux, CNRS. The team analysed high-resolution spectra of six Gaia FGK Benchmark Stars to establish a robust and reproducible set of infrared absorption lines, addressing the current limitations in infrared measurements caused by uncertainties in atmospheric models and data quality. By rigorously evaluating line depth, saturation, blending, and profile agreement, they identified key transitions, notably of alpha-elements and some neutron-capture species, offering a transparent framework for validating near-infrared lines as observational capabilities and theoretical models continue to advance.

Scientists are refining techniques to analyse starlight, bringing us closer to understanding the composition of distant worlds. Accurate chemical fingerprints from stars are vital for tracing the origins of planets and even the building blocks of life. Recent advances in infrared spectrographs have spurred significant progress in understanding atomic and molecular line lists and the models used to interpret stellar atmospheres.

While determining the abundance of elements in stars using optical light has benefited from extensively tested data and modelling, infrared measurements still present greater challenges due to uncertainties in atmospheric models and the quality of available atomic data. Synthetic spectra were generated using established stellar parameters, and each atomic transition was rigorously assessed across all six stars.

This evaluation considered line depth, the extent of line saturation, the presence of blending from other spectral features (purity), and the consistency between observed and synthetic line profiles. The result is a set of robust transitions, spectral fingerprints, that remain consistent regardless of the specific stellar characteristics within the sample.

Lines from alpha-elements like magnesium, silicon, and calcium, alongside several iron transitions, consistently met the stringent criteria for reliability. Notably, among the elements that capture neutrons, only strontium provided lines that consistently satisfied the requirements for accurate abundance determination. This study goes beyond identifying useful lines; it establishes a fully quantitative, multi-criteria framework for validating near-infrared spectral lines.

This framework will be invaluable as laboratory data, stellar atmosphere models, and instrumentation continue to improve, particularly in the era of powerful new telescopes like the James Webb Space Telescope and upcoming facilities such as MOONS. Spectra covering the Y, J, and H bands (9800, 18000 Å) were obtained, focusing on a wavelength range where infrared measurements often exhibit larger uncertainties than optical ones.

Synthetic spectra were generated utilising the established stellar parameters for each benchmark star, enabling a direct comparison between observed and modelled line profiles. Each spectral transition underwent a rigorous, quantitative evaluation across all six stars, assessing line depth, saturation, and the degree of blending with neighbouring features.

This multi-faceted approach ensured that only demonstrably unblended lines were considered for inclusion in the final list. A key methodological innovation involved a four-criteria framework for line validation, assessing lines for depth, saturation, blending (purity), and the agreement between observed and synthetic profiles. This framework prioritises transparency and reproducibility, moving beyond traditional differential abundance methods.

By anchoring the analysis to laboratory-derived atomic data from the Vienna Atomic Line Database (VALD3; Ryabchikova et al0.2015), the work avoids reliance on empirically calibrated line lists, such as those from the APOGEE survey, which may not be universally applicable. Lines of alpha-elements, magnesium I, silicon I, and calcium I, consistently satisfy stringent robustness criteria established for accurate abundance determination.

Several iron I transitions also met these criteria, indicating their reliability for spectroscopic analysis. Signal-to-noise ratios reached 659.7, 470.4, and 359.5 for α Centauri, α Bootis, and β Hydri respectively, ensuring high data quality throughout the study. Effective temperatures for the sample stars range from 3904 K to 6582 K, with surface gravities spanning 1.06 dex to 4.44 dex, and metallicities from -0.55 dex to 0.02 dex.

This broad parameter space was crucial for identifying lines consistently valid across diverse stellar environments. The quantitative framework employed assessed line depth, saturation, blending (purity), and agreement between observed and synthetic line profiles. Lines were subjected to a multi-step filtering process, with only those passing all criteria being designated as robust.

Among neutron-capture species, strontium II uniquely provided lines consistently meeting the established requirements. Observations conducted with CRIRES+ yielded exposure times ranging from 2.9 seconds to 362.9 seconds, optimised for the spectral resolution of approximately 100000 with a 0.2 arcsecond slit. Stellar parameters were derived from angular diameters, bolometric fluxes, and evolutionary models based on Gaia data, ensuring a solid foundation for synthetic spectrum generation. The study highlights the importance of a transparent and reproducible framework for near-infrared line validation, particularly as laboratory data, stellar atmosphere models, and instrumentation continue to advance.

Establishing robust infrared spectral indicators for precise stellar composition analysis

Scientists painstakingly building consensus around fundamental measurements is a hallmark of mature fields, and increasingly vital as astronomy embraces ever more complex data. This work, establishing a rigorously vetted set of infrared absorption lines in benchmark stars, exemplifies that process. For years, stellar abundance studies have relied heavily on optical spectra, benefiting from decades of careful line list development.

Infrared observations, however, have lagged, hampered by the sheer complexity of the spectra and the relative scarcity of precisely measured laboratory data to underpin accurate modelling. What distinguishes this study isn’t identifying robust spectral features, magnesium, silicon, calcium, and a handful of iron lines consistently emerge as reliable indicators, but the methodical framework employed to achieve it.

By demanding consistency across stars with differing properties, and applying multiple quantitative criteria, the researchers have created a foundation for reproducible analysis. This is particularly important as new infrared spectrographs come online, promising a wealth of data but also demanding equally robust calibration tools. Limitations remain, of course.

The focus on a relatively narrow range of spectral types means the line list may not be universally applicable. The scarcity of reliable neutron-capture species lines highlights ongoing challenges in understanding the formation of heavy elements in stars. Future work will undoubtedly expand the scope of this analysis, incorporating more stellar types and pushing towards even greater precision. The real legacy may lie in the demonstrated methodology, a template for building confidence in infrared abundance determinations and unlocking the full potential of this powerful observational window.