EDP Sciences logo
Subscriber Authentication Point
Search
Menu
Home All issues Volume 675 (July 2023) A&A, 675 (2023) A154 Full HTML
Press Release
Open Access
Issue
A&A
Volume 675, July 2023
Article Number A154
Number of page(s) 28
Section Stellar structure and evolution
DOI https://doi.org/10.1051/0004-6361/202245650
Published online 19 July 2023

© The Authors 2023

Licence Creative CommonsOpen Access article, published by EDP Sciences, under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

This article is published in open access under the Subscribe to Open model. Subscribe to A&A to support open access publication.

1. Introduction

We find ourselves amidst a scientific revolution: gravitational wave (GW) observatories will soon be detecting black hole (BH) mergers as frequently as once per day. To interpret these events, we need to comprehend massive stars in low-metallicity environments (Abbott et al. 2020). This is also crucial for other fields of astrophysics, including feedback processes (e.g. Doran et al. 2013), star formation, interstellar medium (ISM) physics, supernovae (SNe), and cosmology. To enable progress in these research areas, we need to uniformly sample the relevant parameter space for massive OB stars, including spectral type, luminosity class, and metallicity (Z).

The Hubble Space Telescope (HST) has dedicated 1000 orbits to the Director’s Discretionary Time project Ultraviolet Legacy Library of Young Stars as Essential Standards (ULLYSES; Roman-Duval et al. 2020)1, making this the largest HST programme ever conducted. ULLYSES is compiling an ultraviolet (UV) spectroscopic Legacy Atlas of about 250 OB stars in low-Z regions2. Due to their proximity, the Large and Small Magellanic Clouds (LMC and SMC) are the best low-Z laboratories for massive star studies, with respectively 50% and 20% Z⊙. They are ideal for studying spatially resolved populations of low-Z massive stars to make a leap towards understanding the early Universe. As a pilot study, several stars at sub-SMC metallicities (in Sextans A and NGC 3109; ∼10% Z⊙) are also included. The aim is to uniformly cover all spectral sub-types from O2 to B9 and to observe all luminosity classes with spectral types O2−B1.5 for both LMC and SMC metallicities, a total of ∼250 stars (Figs. 1 and 2).

thumbnail Fig. 1.

Positions of the ULLYSES/XShootU sources in the LMC (left) and SMC (right). Yellow dots are O-type stars, red diamonds are B-type stars, and blue squares are WR and WR-like Of/WR ‘slash’ stars. We note that the two images have different spatial scales. This figure was made with the Aladin Sky Atlas (Bonnarel et al. 2000); the background consists of DSS2 colour images.

thumbnail Fig. 2.

Distribution of the spectral types in ULLYSES. For the ten known binaries in the sample, only the primary component is accounted for. The category labelled ‘W’ includes WR and WR-like ‘slash’ stars.

Although massive stars emit the bulk of their light at UV wavelengths, the optical region remains the cornerstone of spectroscopic analysis studies. The UV regime is very useful for determining the iron (Fe) abundance and obtaining information on wind parameters, such as the terminal velocity (v∞). The optical regime is crucial for determining the basic stellar parameters, such as effective temperature (Teff), surface gravity (log g), and abundances (Hillier 2020; Simón-Díaz 2020; Brands et al. 2022) Key information on wind clumping and mass-loss rates (Ṁ) only become reliable when optical and near-infrared (NIR) observations are added. Knowledge of the NIR regime is also critical for observations with instrumentation on the James Webb Space Telescope (JWST) and the extremely large telescopes, which will predominately shift our focus to longer wavelengths. Despite the great potential of ULLYSES to transform our knowledge of massive stars, this Legacy dataset is not complete without observations in the optical and NIR regimes. Thus, the X-Shooting ULLYSES (XShootU)3 project was conceived to obtain complementary high-quality spectra of the ULYSSES targets with X-shooter at the European Southern Observatory (ESO) Very Large Telescope (VLT; Vernet et al. 2011).

The optical Large VLT-FLAMES Survey of massive stars (PI: S.J. Smartt) and its successor, the VLT-FLAMES Tarantula Survey (VFTS; PI: C.J. Evans), tackled many science questions, including the Z dependence of stellar wind mass-loss rates (Mokiem et al. 2007b) and the rotation velocities of massive stars (Ramírez-Agudelo et al. 2013). The surface nitrogen (N) abundance of most massive stars in the Magellanic Clouds seemed to be consistent with theoretical predictions, but a significant fraction of stars (20%−40% depending on sample and metallicity) showed chemical enrichment that is either too strong or too weak (e.g. Hunter et al. 2008a; Przybilla et al. 2010; Maeder et al. 2014; Grin et al. 2017).

The absolute mass-loss rates of massive OB and Wolf-Rayet (WR) stars are still uncertain (e.g. Sundqvist et al. 2019; Sander et al. 2020; Ramachandran et al. 2019; Marcolino et al. 2022; Rickard et al. 2022). According to evolutionary models, the bulk of the mass loss could occur during the B supergiant phase rather than during the preceding O-star phase (Groh et al. 2014). The Z dependence of mass-loss behaviour in this cooler regime is highly complex, involving various mass-loss discontinuities as a function of temperature (bi-stability jumps; Petrov et al. 2016), and is critical in predicting BH masses as a function of Z (Belczynski et al. 2010) as well as GW mergers (Kruckow et al. 2016).

The combined UV and optical XShootU project was motivated to address these science questions as well as a large variety of additional questions concerning massive stars at low Z. The project will derive accurate stellar and wind parameters, such as effective temperatures, luminosities, gravities, abundances, and mass-loss rates. This will establish whether mass-loss rates decrease with lower Z, as predicted (Vink et al. 2001; Kudritzki 2002) and empirically supported for relatively small LMC and SMC VLT-FLAMES survey samples (Mokiem et al. 2007b; Ramachandran et al. 2019). Moreover, feedback parameters involving wind momenta, wind kinetic energy, and ionising fluxes are key ingredients for building the next generation of spectral population synthesis models that may be applied to extra-galactic surveys, such as CLUES (Sirressi et al. 2022), the COS Legacy Spectroscopic SurveY (CLASSY; Berg et al. 2022), the HST spectroscopic survey of star-forming galaxies in the local Universe, and future projects. Bright early-type stars are also excellent probes of ISM conditions (van Loon et al. 2013). We expect many spin-off projects that use XShootU and ULLYSES data, including the derivation of the extinction law for which X-shooter’s wide spectral range is particularly useful.

In this work we present the XShootU project. The science requirements are described, as are initial results on data reduction and data analysis processes. We show how the legacy spectroscopic dataset of ULLYSES and XShootU can increase our knowledge of massive stars at low Z. The organisation of the XShootU collaboration is described in the appendix. Already published data (Table B.2, plotted in Fig. 3) may naively suggest that properties of LMC and SMC stars are known, but these pre-ULLYSES results have been derived for relatively small and heterogeneous datasets, and gaps are evident. To make matters worse, spectral analyses that derive stellar properties have also been heterogeneous, as different authors have not only used different tools, distances, and baseline abundances, but also different wavelength ranges.

thumbnail Fig. 3.

HR diagrams of the ULLYSES SMC and LMC targets. Stellar parameters are based on contemporary literature from Table B.2 (filled symbols) or spectral type calibrations (open symbols). For known binary and multiple systems, only primaries are indicated. Calibrations used are Doran et al. (2013) for O-type stars in both galaxies, Dufton et al. (2019), Trundle et al. (2004), and Trundle & Lennon (2005) for SMC B-type stars, and Dufton et al. (2018), Garland et al. (2017), McEvoy et al. (2015), and Urbaneja et al. (2017) for LMC B-type stars. Evolutionary tracks (solid lines) and isochrones (dotted lines) for non-rotating massive stars at 0.5 Z⊙ and 0.2 Z⊙ are from Brott et al. (2011), supplemented by tracks for very massive stars in the LMC from Köhler et al. (2015).

To make progress, not only do the spectroscopic datasets need to be uniform – as provided by ULLYSES and XShootU – but so does the spectral analysis approach. To give an example, determining the Ṁ − Z relationship not only requires accurate mass-loss rate determinations, but also reliable stellar parameters, such as luminosities, to compare the Ṁ from one star with one set of stellar properties in one particular galaxy to the mass-loss rate from another star in another galaxy. A uniform data and analysis approach is at the heart of the XShootU project.

2. XShootU science requirements

In order to build better population synthesis models of massive stars in low-Z environments, such as those at high redshift studied with JWST, we require (i) more complete spectral libraries and (ii) more reliable stellar evolution models for low-Z stars. The former involves the construction of more accurate model atmospheres, but the latter implies a better handle on the behaviour of wind mass loss over a multi-dimensional parameter space, including Z. The key line driver of the inner winds that sets Ṁ of massive OB stars is iron (Fe), while intermediate mass elements such as C, N, and O dominate the outer winds, setting the terminal velocity (Vink et al. 1999; Puls et al. 2000). While high redshift galaxies may have different [α/Fe] ratios compared to local low-Z galaxies, non-solar [α/Fe] ratios should have very little impact on the expected mass-loss rate, as long as one correctly interprets low Z as having low Fe contents (see for instance Table 5 in Vink et al. (2001) for conversions between O and Fe).

One potential concern is whether the local low-Z LMC and SMC at 0.5 Z⊙ and 0.2 Z⊙ are sufficiently metal-poor to gain insight into low-Z stellar evolution in high-redshift galaxies. In order to make the case that the SMC indeed has a sufficiently low Fe-contents to provide key insights into the early Universe, we showcase a number of MESA stellar evolution models (see Appendix A.5) in Fig. 4. The plot indicates that a rapidly rotating massive star at LMC metallicity still shows classical redwards evolution, just like in the Milky Way, but that already the one-fifth solar SMC metallicity is sufficiently low to undergo bluewards chemically homogeneous evolution (CHE), similar to even lower-Z stars at one-tenth solar. The Galactic model loses as much as a third of its initial 50 M⊙ mass already on the main-sequence, while the SMC and lower-Z models lose of the order of 10% or less. Moreover, while the Galactic and LMC models completely spin down during the main sequence (the Galactic model drops below the minimum observable v sin i value of 100 km s−1 after 3 Myrs, while the LMC star can delay this to 4.5 Myrs), the SMC and lower-Z models hardly spin down at all. In fact, the SMC and lower-Z model evolve towards critical rotation, rather than away from it. In other words, the SMC is an ideal test-bed for gaining an understanding of the physical difference between the high-Z and the low-Z Universe.

thumbnail Fig. 4.

Main-sequence MESA stellar evolution models of a rapidly rotating (v sin i = 550 km s−1) 50 M⊙ star for a range of metallicities. The Galactic and LMC models show traditional redward evolution, while the SMC and even lower-Z (one-tenth solar) models start to show blueward chemical homogeneous evolution. The models employ Vink et al. (2000, 2001) mass-loss rates and assume a moderate amount of core overshooting, with a value of αov of 0.335, similar to the Brott et al. (2011) models.

The results displayed in Fig. 4 may naively give the impression that stellar evolution is already well understood, but this is not the case, and the stellar evolution and population synthesis models are only as good as the input physics. In this parameter space those are predominately given by the assumed amounts of interior mixing and wind mass loss. It is commonly assumed that the only parameters setting the mass-loss rate are the stellar luminosity and the metallicity, but in the oft-used mass-loss recipe of Vink et al. (2000, 2001) parameters such as stellar mass, and effective temperature – including the B supergiant regime below the bi-stability jump – also play a crucial role. Therefore, in order to make progress on the accuracy of stellar evolution models at low Z we firstly require large samples of wind parameters offered by the ULLYSES sample. Secondly, in order to test these in different parts of the Hertzsprung-Russell (HR) diagram, the underlying stellar parameters also need to be robust. Thirdly, in order to test the role of rotational mixing for a range of metallicities we require stellar abundances.

Starting with the third requirement, massive stars undergo H-burning via the CNO cycle, and in the first instance the core nitrogen (N) abundance is expected to increase by an order of magnitude at the expense of carbon (C; Brott et al. 2011; Ekström et al. 2012). Mixing can bring enhanced N to the surface, which is especially relevant for testing the physics of rotational mixing in stellar evolution models. Factors of 2–10 in N enhancement and C depletion are realistically measurable from UV and optical spectroscopy as discussed in Sect. 4. For the stellar parameters, effective temperatures need to be accurate to within 5–10%, which is routinely achieved in non-local thermodynamic equilibrium model (NLTE) atmosphere modelling. More cumbersome is the estimated log g that determines the spectroscopic mass. In Sect. 4 we show that in order to be able to derive accurate log g the UV alone does not suffice, and optical Balmer lines are mandatory (see below). Arguably the least well-constrained parameter is the wind mass-loss rate. While UV P Cygni lines offer relatively accurate values of the terminal wind velocity, ∼10% (Prinja et al. 1990), uncertainties in empirical mass-loss rates are about an order of magnitude due to the roles of, respectively, micro-clumping and macro-clumping (Fullerton et al. 2006; Oskinova et al. 2007; Sundqvist et al. 2018) Clearly, such huge uncertainties are not acceptable when building reliable stellar evolution and populations synthesis models. From our experience in stellar modelling, such as the experiments performed in Fig. 4, we conclude that we need the mass-loss rate to be accurate to 0.3 dex.

As the mass-loss rate is a multi-variate function of stellar parameters, such as L, M, and Teff, the accuracy requirements on the stellar parameters need to be at least as good as those for the mass-loss rate. Accuracies on Teff are easily within 10%, though precisions of 1 kK are sometimes quoted. Similarly, log L precisions of 0.1 dex are feasible. However, the real culprit is the stellar mass, M, which can be obtained from log g spectroscopically but which has a long history of uncertainty, culminating in systematic differences between these spectroscopic masses and evolutionary masses of the order of a factor of ∼2 (Herrero et al. 1992). In Sect. 4 we show that the optical regime is absolutely critical to measure log g.

3. XShootU data description

3.1. Target selection

The first objective of the XShootU project is to create a homogeneous legacy atlas of similar quality and scope as that of ULLYSES. The target sample contains 132 LMC stars, 106 SMC targets, and 6 very-low-Z stars in Sextans A and NGC 3109 (Roman-Duval et al. 2020; see Table B.1 of this paper).

Most ULLYSES targets are O-type stars (154), but B-type stars (72), and WR/slash stars (18) are also included. Figure 1 displays the positions of the targets on the sky, and Fig. 2 shows the distribution of spectral types. The SMC targets have masses in the range of 10 − 60 M⊙, whereas the LMC targets have masses in the range of 15 − 150 M⊙. A subset of the ULLYSES targets have previously been spectroscopically analysed (see Table B.2). These heterogeneous pre-ULLYSES data are presented in the HR diagram in Fig. 3 (filled symbols). Estimated parameters for targets lacking contemporary analyses are also shown (open symbols).

XShootU obtained a complementary dataset over the optical to NIR wavelength range for all ULLYSES targets that have not previously been observed with X-shooter. This resulted in a sample of 1294 LMC stars and 103 SMC stars. In addition, three very low-Z stars were included in the sample.

The ULLYSES and XShootU datasets are not taken simultaneously in time, although, apart from a few exceptions, the vast majority of ULLYSES sources are not known variables. In reality, most stars are variable to some level, so care still needs to be taken when interpreting the data, but we do not anticipate this to be a massive issue. Existing ESO Science Archive Facility data are part of both spectroscopic and time-dependent aspects of XShootU. Over half of the ULLYSES targets have no previous high-quality optical spectra. Several have been observed with UVES (15%) and/or FLAMES (∼50%), but the wavelength coverage of these FLAMES data is limited. Using (limited) time-sequence data, we searched for binary signatures, and in some cases be capable of disentangling spectra of multi-component systems (Mahy et al. 2020).

3.2. XShootU observing strategy

3.2.1. Wavelength coverage

ULLYSES obtained moderate resolution spectra of OB stars with selected wavelength settings of the COS G130M, COS G160M, STIS E140M, COS G185M, and STIS E230M gratings in the far- and near-UV during HST cycles 27 − 29. In order to complement this UV range, similar quality optical/NIR spectroscopy was carried out with the X-shooter instrument. This slit-fed (11″ slit length) spectrograph provides simultaneous coverage of the wavelength region between 300 − 2500 nm, divided into three arms; UVB (300 ≲ λ ≲ 500 nm), VIS (500 ≲ λ ≲ 1000 nm), and NIR (1000 ≲ λ ≲ 2500 nm). X-shooter’s wide wavelength coverage made it the instrument of choice for the purpose of building an optical-NIR legacy dataset.

3.2.2. Spectral resolution

The X-shooter slit widths were chosen to obtain a spectral resolution of R = 5000 − 10 000, required for estimating the stellar parameters. Each target was observed with a set of (UVB, R = 6700), (VIS, R = 11 400), and (NIR, R = 8100) slit widths, matching also the average seeing conditions on Paranal. The slit position angle was set by default to the parallactic angle, but when necessary a fixed position angle on the sky was used to optimally prevent nearby sources from entering the slits.

Figure 5 illustrates that although the spectral resolution is only medium, the determination of surface abundances, including nitrogen (N), carbon (C), and oxygen (O), should be feasible. It is the projected rotational velocity of some of the stars that will limit such studies. The higher v sin i, the broader the lines, which become challenging to identify at very high v sin i. Even with a S/N around 100 (see the justification below) most lines would be undetectable at high v sin i. On the other side of the distribution, lines are only partially resolved at low v sin i. Consequently, the determination of accurate low v sin i values is not feasible, and additional higher spectral resolution data are needed for this subset (see Appendix A.11 for a description of auxiliary Magellan/MIKE data), although the combined analysis of optical and UV lines can partially alleviate this limitation. The bottom panel of Fig. 5 illustrates the effect of rotational broadening of C III 1176, a line complex relevant for the determination of v sin i (e.g. Bouret et al. 2013). Above 100 km s−1, the components of the multiplet are blended, while at lower rotational velocities they are resolved individually.

thumbnail Fig. 5.

Effect of spectral resolution and rotational velocity on three sets of optical lines classically used to determine C, N, and O abundances. In each panel, the initial CMFGEN model has Teff = 31 000 K, log g = 3.6 and 0.2 Z⊙. The model is degraded to a resolution of either 6000 (typical for our X-shooter UVB spectra) or 80 000 and further convolved with three rotational velocities (20, 100, and 300 km s−1). No additional macro-turbulent broadening is considered. Also plotted is the UV C III 1176 line. Here the spectral resolution is that of the STIS E140M grating (R ∼ 45 000).

3.2.3. Signal-to noise ratio

Some VLT instruments provide higher spectral resolution in certain wavelength regimes (e.g. UVES), but to build a homogeneous database with a wide spectral coverage could only be achieved with X-shooter. In addition to the wide wavelength coverage, a high signal-to-noise ratio (S/N) in addition to sufficient spectral resolution are essential to determine the fundamental stellar parameters and the abundances for various temperature regimes populated by OB and WR stars. For the preparation of the X-shooter proposal, we estimated the required S/N, experimenting on a typical SMC mid-O dwarf/giant with CMFGEN model spectra degraded to X-shooter’s spectral resolution. We found that the determination of basic stellar parameters such as Teff and log g became prohibitive for quantitative interpretation if the S/N drops below 100 per resolution element. To ensure the maximal scientific return of XShootU, we achieved a S/N of > 100 in the continuum in the UVB and VIS for all Magellanic Cloud targets.

3.3. Data reduction

A detailed description of the data reduction is provided in a paper associated with Data Release 1 (DR1; Sana et al., in prep., hereafter XShootU II). Here, we provide a brief summary, focusing on the UVB and VIS spectra. The data reduction of the NIR spectra requires additional efforts and will become part of DR2.

The initial data reduction was performed using the ESO X-shooter pipeline v3.5.0 (Goldoni 2011). The pipeline carried out the standard steps of bias, flat, wavelength calibration, spectral rectification, cosmic ray removal, sky subtraction, flux calibration, and extraction of a 1D spectrum. The wavelength calibration was performed using a physical model, that is to say, the transformation from pixel to lambda space was optimised through the analysis of a multi-pinhole ThAr (UVB, VIS) or pen-ray (NIR) lamp frame. The predicted positions of the lines were fitted using a 2D Gaussian to recover the actual positions on the frame.

The pipeline-reduced data were subsequently flux calibrated using a set of six standard stars observed during the same or adjacent nights. We found that the stellar models used by the public pipeline (Moehler et al. 2014) resulted in small (of the order of a few percent) changes in the Balmer line profiles depending on the standard star. This potentially impedes accurate log g measurements. In addition, the spectral energy distribution for some of the standard stars could be optimised. We decided to use new stellar models and new fit points to derive the response, starting with models used by HST for their fundamental flux standards GD 71 and GD 153 (Bohlin et al. 2020)5. We then reduced observations of the other five standard stars taken between October 2020 and April 2021 as if they were science objects, with the response determined by close-in-time observations of GD 71. Those spectra were then co-added to create high S/N spectra that were used to derive improved stellar models. The XShootU spectra were obtained with narrow slits. To obtain absolute flux calibrated spectra, corrections were applied for slit losses due to seeing and image quality across the detector and by re-scaling to existing photometry. The achieved accuracy is typically better than 5%.

Telluric correction was performed using the molecfit tool v3.0.3 (Smette et al. 2015; Kausch et al. 2015) for the VIS arm and generally leads to good results. For the Magellanic Cloud targets, we fitted the atmospheric model directly to the science spectra, as the S/N on the continuum is high enough to ensure a better correction than using a telluric standard star to compute the model. The regions with very deep O2 telluric absorption lines at ∼760 nm and sometimes the one at ∼690 nm are poorly corrected and the correction of the H2O bands at ∼950 nm leaves strong residuals. The correction from telluric lines around the [OI] 6300 Å line is always good.

Results of the data reduction are shown in Fig. 6. Here a sample of reduced X-shooter spectra is presented to highlight a sequence from the earliest to the latest spectral types for supergiant and dwarf targets. The spectra shown are single-epoch in order to avoid confusion in co-added spectra due to potential variability. Telluric correction (grey regions) and proper cosmic-ray removal was not performed for this plot, but will become part of DR1 (XShootU II). An example of an O4 supergiant spectrum is shown in Fig. 7 on a improved scale, focused on wavelength regions in which telluric corrections are not needed.

thumbnail Fig. 6.

Reduced X-shooter spectra for a range of spectral types of single-star supergiants (top) and dwarfs (bottom). For illustration purposes, the flux of each spectrum was divided by its mean value and an arbitrary offset was added. The grey regions correspond to the UVB-VIS common wavelength coverage (∼ 5500 Å), a gap due to bad pixel masking (∼ 6360 Å), and telluric absorption. Minor manual treatment to remove strong cosmic rays was performed.

thumbnail Fig. 7.

X-shooter/UBV spectrum of Sk −67° 167 (O4 Inf+) in the LMC, including zoomed-in views of key spectroscopic diagnostics.

4. Multi-wavelength analyses

In Sect. 2 we show that the low-Z environment of the SMC can be considered rather characteristic of the early Universe, with low mass-loss rates, and the potential for rapid rotation and blueward evolution, while the LMC sample is more characteristic of today’s Universe, with higher mass-loss rates, slower rotation, and classical redward stellar evolution. In reality, the situation is more complex, as the mass-loss rates is a function of Ṁ = f(Z,L,M,Teff, v sin i), which implies we need to obtain stellar and wind parameters over a large parameter space, including not only the O-star regime, but also the B supergiant regime, where the bi-stability jump may increase mass-loss rates (Vink et al. 1999), or not (Björklund et al. 2021). Moreover, stellar evolution models depend on interior mixing, and stellar abundances can be utilised to test the efficiency of (rotational) mixing.

4.1. Diagnostics in the UV and optical range

In the optical range, He/H abundances can only be determined from the optical since H/He lines in the UV are dominated by strong interstellar features (e.g. Lyman alpha) so the Pickering-Balmer lines in the optical are critical for the He/H ratio.

Abundances of C, N, and O can in principle be determined from the UV range only (e.g. Bouret et al. 2003, 2013) but most lines are also sensitive to winds, especially as one moves away from the main sequence. Figure 8 shows an example where winds are sufficiently weak for such a determination. The plot also highlights that the optical range contains more lines from these elements, and these lines depend far less on wind properties than those in the UV. Using more lines reduces the systematic uncertainties in the determinations. Figure 8 highlights that it is still challenging to obtain a perfect fit for all lines of the same element, but the availability of more lines helps in identifying potential shortcomings in the atmosphere models. It also allows for a better determination of errors associated with abundance determinations, which is crucial for interpreting stellar evolution predictions of interior mixing. A full error determination will follow in a dedicated paper, but we could already say that typical error bars are 15–30%, sometimes up to 50%, with these two datasets combined (see Bouret et al. 2021), easily satisfying our science requirements.

thumbnail Fig. 8.

Comparison between the observed HST (top) and X-shooter spectrum (black line) of selected lines of C, N, and O for AzV 327 (O9.5 II-Ibw) with two models (coloured lines). The light blue line is for solar-scaled abundances (factor of 1/5), while the red model has the following scaling: C abundance decreased by a factor of 3.8, N abundance increased by a factor of 4.5, and O abundance decreased by a factor of 1.6. The models were computed with the NLTE CMFGEN (Hillier & Miller 1998) atmosphere code.

Another key aspect of combining the ULLYSES and XShootU datasets is that it allows stellar and wind parameters to be derived self-consistently using both optical and UV diagnostics, which was lacking in surveys such as VFTS. For O-type stars, there are no conclusive diagnostics in the UV to derive effective temperatures and gravities. The wind profiles of O IV, O V, and N IV impose a minimum effective temperature, but they are also sensitive to the mass-loss rate and clumping properties. Figure 9 highlights which parameters can be determined from specific UV as well as optical lines for an LMC O supergiant.

thumbnail Fig. 9.

UV (top) and optical (bottom) spectrum of the LMC star Sk-67 167 (O4 Inf+). The UV spectrum consists of STIS E140M observations taken as part of the ULLYSES project and archival FUSE data. The optical spectrum was obtained with X-shooter. A selection of diagnostics for stellar and wind parameters are highlighted. Note that these diagnostics can vary with spectral type. The ticks at the bottom of the UV spectra mark the position of interstellar lines.

As 30% of the sample involves B stars rather than O-type stars, we also show Fig. 10 highlighting which parameters can be determined from the UV versus optical part for B supergiants (Crowther et al. 2006; Firnstein & Przybilla 2012; McEvoy et al. 2015). The combination of UV and optical spectra is even more powerful for constraining the physical properties of B-type stars. As in the case of O-stars their UV spectrum alone does not contain diagnostics for gravity. Effective temperature could to first order be constrained by comparing the CII/CIII lines and the FeIII/Fe IV line forests, although the C lines are also sensitive to mass loss, and Fe transitions depend also on log g. The optical range offers cleaner Teff diagnostics from the ionisation balance of SiII/SiIII/SiIV (secondarily, the comparison of HeI and MgII) and gravity (e.g. Hγ and the higher Balmer lines). The numerous metallic lines in the optical can be used to determine abundances and in the case of the strongest transitions (e.g. SiIII) micro-turbulence, and projected rotational velocity. The joint UV + optical range offers several mass-loss rate and clumping diagnostics, with the SiIV doublet being the best wind velocity indicator for early B supergiants.

thumbnail Fig. 10.

UV (top, middle) and optical (bottom) spectrum of the SMC star Sk191 (B1.5 Ia). The UV spectrum consists of FUSE, STIS E140M, and STIS E230M observations compiled as part of the ULLYSES project. The optical spectrum was obtained with X-shooter. Similarly to Fig. 9, interstellar transitions and a selection of stellar and wind diagnostics are highlighted. In addition, metallic lines that can be used to measure abundances are marked in purple.

In order to further quantify the need for optical X-shooter spectra in spectroscopic analyses, we present an example analysis for an O8 III SMC giant in Fig. 11. The figure shows both the UV part of the spectrum and some Hydrogen Balmer lines that are routinely utilised to derive log g values. It can easily be seen that while high and low log g model values reproduce the UV spectra equally well, the optical is critical for accurate log g determination. While the complexity of the spectroscopic analysis is beyond the scope of this paper, the key point is that UV-only fits yield very poorly constrained surface gravities, which result in enormous uncertainties on spectroscopic masses.

thumbnail Fig. 11.

O giant (AV186, O8.5III). The best fitting model is for T = 33 kK and log g = 3.4. Note that while a ten times higher mass star with log g = 4.4 would be indistinguishable in the UV, it would completely fail to reproduce the optical Balmer wings. The model was computed with the PoWR (Sander et al. 2017) NLTE code.

In addition to the uncertainty in surface gravity, it is also appropriate to mention that for stars with strong winds, such as supergiants, the Hα Balmer line is a key mass-loss and clumping factor diagnostic. When only accounting for micro-clumping and the UV part of the spectrum, Fullerton et al. (2006) showed that clumping factors were uncertain by factors of up to a hundred, and mass-loss rate reductions could easily be an order of magnitude. Only when accounting for the optical Hα line and macro-clumping (see Appendix A.4), Oskinova et al. (2007) showed that mass-loss rate uncertainties were significantly smaller, by a factor of 2 or so, and from the additional optical Hα line, clumping factors are usually estimated to be lower, of the order of 6–8 (e.g. Ramírez-Agudelo et al. 2017).

4.2. Spectroscopic analysis tools and procedures

XShootU is coordinating spectral modelling efforts for massive stars on a world-wide scale never witnessed before in the massive-star community. Before we can scale-up the analysis to hundreds of massive stars with hugely varying spectral and wind properties over the entire hot part of the HR diagram, it is paramount that codes and analysis techniques are tested and compared as a function of stellar parameters and metallicity.

The spectral analysis of massive stars is rather intricate due to the highly NLTE conditions in their turbulent, supersonic atmospheres. Over the past few decades, a number of highly complex, yet successful, model atmosphere codes have been developed, for example CMFGEN (Hillier & Miller 1998), PoWR (Hamann & Gräfener 2003; Sander et al. 2015), and FASTWIND (Santolaya-Rey et al. 1997; Puls et al. 2005). Although these codes have previously been applied to various sets of observations, only more recently have they been used for larger samples (e.g. Ramírez-Agudelo et al. 2017; Sabín-Sanjulián et al. 2017) due to the efficiency of numerical methods (applying certain physical approximations), and efficient spectral automated analysis tools including genetic algorithms and grid-based χ2 approaches.

At virtual and on-site Lorentz workshops in 2021 and 2022 (and various additional virtual XShootU meetings) preliminary comparisons of analyses with the various NLTE codes were performed by modelling subsets of O-stars. Agreement was reached on a common methodology for performing the spectroscopic analyses within the XShootU Project.

This recommended procedure is summarised as follows:

  1. Use the same reference value for the distance moduli (DM) of the LMC and the SMC. We adopted DM(SMC) = 18.98 mag (Graczyk et al. 2020) and DM(LMC) = 18.48 mag (Pietrzyński et al. 2019).

  2. Adopt the same photometry (U, B, V, J, H, KS, as minimum) for each star (Table B.1). Optical photometry is from a variety of sources, whereas NIR photometry is from VISTA VMC (JKs; Cioni et al. 2011), 2MASS (JHKs; Cutri et al. 2003), or 2MASS 6X (JHKs; Cutri et al. 2012). H-band photometry is omitted if JKs values are discrepant between 2MASS and VMC owing to photometric variability or crowding.

  3. Adopt the same literature source for the bolometric correction. The relation for bolometric correction as a function of Teff and metallicity from Lanz & Hubeny (2003) is adopted for now. This relation may be updated in the course of this project.

  4. Adopt the same reddening approach. Key references for the Milky Way foreground are Fitzpatrick et al. (2019), over earlier works by Seaton (1979), Cardelli et al. (1989), though Galactic foreground extinction is modest towards the Magellanic Clouds. For the Magellanic Cloud contributions, Gordon et al. (2003) is preferred, recognising earlier contributions from, for example, Howarth (1983) and Fitzpatrick (1986) to UV laws in the LMC, and for the SMC from, for example, Prevot et al. (1984) and Bouchet et al. (1985) to the UV and optical/IR, respectively. For the FUSE range, Gordon et al. (2009) and Cartledge et al. (2005) are recommended for the Milky Way, and Magellanic Clouds, respectively.

  5. Adopt the same baseline LMC and SMC abundances. Several abundance ratios in the Magellanic Clouds are notoriously non-solar, which is especially true for CNO in the SMC, and the use of scaled-solar values should be avoided when possible. Thus, the adopted abundance values were derived from an average of different determinations, for example from stars, H II regions, and SN remnants. The recommended values are listed in Tables 1 and 2. For several species, however, we need to agree on default abundances, given the absence of lines in spectral ranges covered by ULLYSES + XShootU. In such cases, scaled-solar values need to be adopted, using 0.5 Z⊙ and 0.2 Z⊙ scaling factors for the LMC and SMC, respectively (Asplund et al. 2009).

  6. Whenever possible, adopt the same description of macro-turbulence for line broadening. The recommendation is to adopt a radial-tangential description of macro-turbulence (e.g. Simón-Díaz & Herrero 2014).

  7. Adopt similar wind clumping implementation. We agreed to use the same parametric description as implemented in CMFGEN (Hillier & Miller 1998) for the standard derivation of the mass-loss rate. Clumping is predominately treated in the ‘micro-clumping’ approximation, assuming a void interclump medium. The volume filling factor, fV, has the following velocity-dependent behaviour:

    (1)

    where fV, ∞ denotes the value at r → ∞. For a void inter-clump medium, the corresponding clumping factor D ≡ fcl is simply the inverse value, implying that . The free parameter vcl is a characteristic velocity varied to describe the clumping onset. More sophisticated descriptions of the properties and nature of clumping can be implemented in the codes (e.g. Oskinova et al. 2007; Hawcroft et al. 2021; Flores & Hillier 2021). These predominately fall under the framework of working group (WG) 4 that focuses on wind structure (see below).

Table 1.

Baseline LMC metal abundances (X/H by number) with respect to Z⊙ from Magg et al. (2022, MBS22).

Table 2.

Baseline SMC metal abundances (X/H by number) adopted with respect to Z⊙ from Magg et al. (2022, MBS22).

The next step is to benchmark the accuracy of stellar parameters derived with the different approaches. For this, a small set of stars was modelled with various NLTE codes. The obtained stellar parameters were compared, allowing us to assess to what extent the physical interpretation depends on the modelling tools applied. In parallel, we also considered bench-marking of the NLTE wind codes (i.e. against one another or against mock data) to perform a direct comparison of synthetic spectra computed for the same input model parameters. Alternatively, we could use a model obtained with one code and fit this model with the other codes. These various ‘bench-marking’ approaches will provide relevant information of systematic differences between codes, analyses tools, and other differences in approach, which will be detailed in a future (benchmarking) paper (Sander et al., in prep.; XShootU IV).

Determination of the wind terminal velocities (v∞) is also taken on. As a global wind parameter, this is indeed an essential input for the models of stellar atmospheres used. The ULLYSES (UV) data are of particular importance for this task as they are rich in resonance transitions of ionised species, which are prime v∞ diagnostics. For stars presenting saturated resonance-line UV profiles, considering that the spectral lines remain optically thick at the distances where the wind reaches its maximum speed, v∞ can be measured directly by measuring the maximum velocity shift of the absorption component of the UV C IV resonance doublet (see e.g. Prinja et al. 1990; Prinja & Crowther 1998). Alternatively, the wind speeds can also be measured by fitting synthetic spectra produced using the Sobolev with exact integration (SEI) method (Lamers et al. 1987). This latest approach is particularly relevant for stars without saturation in their UV resonance-line profile (although it can also be used in the first case mentioned above). As this paper is being written, a significant fraction of the LMC and SMC star sample has been studied with either method. Results concerning the dependence of v∞ with the ambient metallicity or stellar parameters will be presented in dedicated papers (e.g. Hawcroft et al. 2023 XShootU III).

5. Final perspectives

The XShootU project is expected to provide many pieces of data, models, and new physics of massive stars in low-Z environments. It is important to stress that the overarching aim is to provide a high-quality homogeneous optical database that is complementary to ULLYSES. These legacy datasets are critical for a correct interpretation of unresolved high-Z observations with JWST/NIRSpec (Curti et al. 2023; Carnall et al. 2023; De Barros et al. 2019). The next goal is to provide uniformly determined stellar and wind parameters from the combined UV and optical datasets. For this part of the project, it is not only critical to include the correct NLTE physics, but also to test the various spectral synthesis codes and analyses.

A key science aim is to quantify “how” the mass-loss rate declines with decreasing metallicity. This does not necessarily simply imply a determination of a power-law exponent, as the slope may easily vary with stellar parameters or Z itself (Vink et al. 2001; Kudritzki 2002; Sander & Vink 2020; Marcolino et al. 2022; Rickard et al. 2022). In order to obtain an empirical Ṁ = f(Z) relationship, we not only require mass-loss rates and clumping properties, but simultaneously also need to obtain the underlying stellar parameters (Teff, log g, L, and M) as these parameters enable mass-loss properties from a given object in one galaxy to be compared to those from an entirely different object located elsewhere.

The XShootU dataset, coupled with surveys of H II regions in the LMC and SMC such as SDSS LVM6 (Local Volume Mapper), represents a significant opportunity to advance the state-of-the-art in our understanding of how stars shape their environment. LVM is an optical (3600 − 10 000 Å) integral field unit (IFU) spectroscopic survey (R ∼ 4000) of the Milky Way and the Local Group (LMC, SMC, M31, and M33). It will be the first IFU survey to isolate and resolve distinct environments within galaxies and to cover significant portions of the night sky. New population synthesis models – informed by the new physics obtained in the XShootU collaboration – can be used to remove the stellar contribution from LVM observations of star-forming regions, which will enable studies of the ionised gas alone (H II regions and diffuse ionised gas) over the large dynamic size range of LVM, from clusters and clouds (10–50 pc) to the kiloparsec scales of spiral arms, galactic inflows and outflows, and disk dynamics.

One should be aware that different communities refer to Z in different ways. Extra-galactic communities usually work on the basis of nebular oxygen (O) lines, while stellar astronomers are sometimes able to derive Fe abundances of individual stars. It is predominately the Fe abundance that sets the mass-loss rate, while intermediate mass elements such as O set the wind terminal velocity (Vink et al. 1999; Puls et al. 2000). While the [α/Fe] ratio in local dwarf galaxies such as IC 1613 is generally found to be sub-solar (Tautvaišienė et al. 2007; Garcia et al. 2014), Steidel et al. (2016) and Strom et al. (2022) show O to be enhanced in comparison to Fe for galaxies at intermediate redshifts (‘Cosmic Noon’). This is thought to be due to the time delay in the production of Fe from Type Ia SNe with respect to α elements released by massive stars. It may therefore become relevant to consider more detailed abundance patterns than simply scaling all metals with the solar-abundance pattern, that is, to make a clearer distinction between [Fe/H] and α elements for Ṁ calibrations of massive stars, as indicated in Table 5 of Vink et al. (2001). The ULLYSES and XShootU sample will uniquely provide the opportunity to investigate potential differences between [Fe/H] and α elements with non-solar patterns.

When we are eventually able to provide a reliable empirical Ṁ = f(Z) relation, we will be able to compare these findings to theoretical predictions, and this will inform us on how to treat mass loss more reliably in models of stellar evolution – and thereby also in feedback and population synthesis studies – at low Z. It is currently unclear if most of the mass loss takes place in Z-dependent winds, or in Z-independent eruptions or binary interactions. Despite the possibility of significant Z-independent mass loss, there is ample evidence that massive-star evolution in metal-poor environments proceeds very differently from such evolution in our Milky Way. Shirazi & Brinchmann (2012) and Kehrig et al. (2015) found strong nebular He II emission lines in low-Z galaxies, but not in metal-rich ones. Long-duration gamma-ray bursts (e.g. Vreeswijk et al. 2004), super-luminous SNe (e.g. Quimby et al. 2011; Gal-Yam 2012; Chen et al. 2015), and broad-line Type Ic SNe occur preferentially in low-Z dwarf galaxies (e.g. Palmerio et al. 2019), whereas ordinary Type Ic SNe avoid dwarf galaxies.

The spectacular merger of two ∼30 M⊙ BHs observed through the detection of GWs by the LIGO observatory (Abbott et al. 2016a) most likely originated from a system that had an initial Z similar to that of the SMC (Abbott et al. 2016b). This conclusion is inextricably linked to the predicted reduced stellar wind mass loss at low Z (e.g. Vink et al. 2001). Proposed channels that may have led to GW 150914 (and other more recent events) involve mass transfer and common envelope evolution (e.g. Belczynski et al. 2016) and CHE systems (Mandel & de Mink 2016). The CHE model is linked to rapid spin rates, leading to rotationally induced mixing of the stellar interior (Maeder 1987; Yoon & Langer 2005). Predictions of single star evolution have shown this process to be increasingly important at lower Z (Brott et al. 2011) due to lower mass-loss rates, and correspondingly less angular momentum loss. Support for the existence of CHE can be found in the properties of a small fraction of WR stars (Martins et al. 2009; Hainich et al. 2015), but CHE still lacks unambiguous observational confirmation in the O-star regime (see for example Walborn et al. 2004; Abdul-Masih et al. 2019, 2021 and Bouret et al. 2013 for candidates).

The properties of GW events and exotic SNe illustrate the key physics involved: spin rates and rotationally induced mixing, CHE, and wind mass-loss rates. The XShootU sample consists of uniformly estimated spectral parameters of objects previously classified in the literature to varying extents. Further analysis will include, for instance, a determination of an upper limit to the number of CHE stars for which three basic observables need to be fulfilled: (i) a peculiar HR diagram location, as chemically homogeneous stars evolve bluewards instead of redwards, (ii) higher average v sin i, as they are thought to be caused by rapid rotation, and (iii) special chemical-abundance patterns that showcase chemical mixing.

These are just some of the studies that are being prepared, and their results will be published in research articles of the XShootU series. It should be emphasised that some of the analysis is already ongoing, but there is ample space for new parties to join the project. Moreover, the X-shooter data are open to the public, and we also plan to make the higher level data products open to the community at large, as these high-quality data have long-term utility for research projects that may not yet be foreseen.

2

The ULLYSES programme is also compiling high-quality far-UV, near-UV, and optical spectra of young, low-mass T Tauri stars in our Galaxy.

4

These numbers are slightly different as a few archival X-shooter datasets were available.

7

An early version of these tools can be found at https://github.com/orlox/StarStats.jl

Acknowledgments

We thank the ESO support staff for the help in the preparation of the observations, in the scheduling, and for carrying out the observations at Paranal. We also thank the ESO support staff, in particular Carlo Felice Manara. We also thank the ULLYSES teams at STScI, in particular, Julia Roman-Duval and STScI Director Ken Sembach, for executing and enabling the ULLYSES Director’s Discretionary program. Based on observations obtained with the NASA/ESA Hubble Space Telescope, retrieved from the Mikulski Archive for Space Telescopes (MAST) at the Space Telescope Science Institute (STScI). STScI is operated by the Association of Universities for Research in Astronomy, Inc. under NASA contract NAS 5-26555. JSV and ERH gratefully acknowledge support from STFC via grant ST/V000233/1. LPM thanks CNPQ for financial support through grant 307115/2021-6. AW acknowledges the support of UNAM via grant agreement PAPIIT no. IN106922. ADU acknowledges support from NASA under award number 80GSFC21M0002. DMB gratefully acknowledges funding from the Research Foundation Flanders (FWO) by means of a senior postdoctoral fellowship with grant agreement number 1286521N. AACS acknowledges support by the Deutsche Forschungsgemeinschaft (DFG - German Research Foundation) in the form of an Emmy Noether Research Group – Project-ID 445674056 (SA4064/1-1, PI Sander). BK gratefully acknowledges support from the Grant Agency of the Czech Republic (GAČR 22-34467S). The Astronomical Institute in Ondřejov is supported by the project RVO:67985815. CJE gratefully acknowledges support for this work provided by NASA through grant number HST-AR-15794.001-A from the Space Telescope Science Institute, which is operated by AURA, Inc., under NASA contract NAS 5-26555. NSL wishes to thank the National Sciences and Engineering Council of Canada (NSERC) for financial support. AuD acknowledges support by NASA through Chandra Award number TM1-22001B and GO2-23003X issued by the Chandra X-ray Observatory 27 Center, which is operated by the Smithsonian Astrophysical Observatory for and on behalf of NASA under contract NAS8-03060. In addition, AuD acknowledges NASA ATP grant number 80NSSC22K0628. GM acknowledges funding support from the EuŁropean Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 772086). RI gratefully acknowledges support by the National Science Foundation under Grant No. AST-2009412. JMA acknowledges support from the Spanish Government Ministerio de Ciencia e Innovación through grant PGC2018-095 049-B-C22. CJKL gratefully acknowledges support from the International Max Planck Research School for Astronomy and Cosmic Physics at the University of Heidelberg in the form of an IMPRS PhD fellowship. NDK acknowledges support from the National Solar Observatory, which is managed by the Association of Universities for Research in Astronomy, Inc. and funded by the National Science Foundation. MG and FN gratefully acknowledge funding by grants PID2019-105552RB-C41 and MDM-2017-0737 Unidad de Excelencia “María de Maeztu”-Centro de Astrobiología (INTA-CSIC) by the Spanish Ministry of Science and Innovation/ State Agency of Research MCIN/AEI/10.13039/501100011033 and by “ERDF A way of making Europe”. We thank the Lorentz Centre in Leiden for hosting “ULLYSES sets sail: massive star spectroscopy with HST and the ESO VLT”.

References

  1. Abbott, B. P., Abbott, R., Abbott, T. D., et al. 2016a, Phys. Rev. Lett., 116, 241102 [NASA ADS] [CrossRef] [Google Scholar]
  2. Abbott, B. P., Abbott, R., Abbott, T. D., et al. 2016b, ApJ, 818, L22 [Google Scholar]
  3. Abbott, B. P., Abbott, R., Abbott, T. D., & Abraham, S. 2020, ApJ, 892, L3 [NASA ADS] [CrossRef] [Google Scholar]
  4. Abdul-Masih, M., Sana, H., Sundqvist, J., et al. 2019, ApJ, 880, 115 [Google Scholar]
  5. Abdul-Masih, M., Sana, H., Hawcroft, C., et al. 2021, A&A, 651, A96 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  6. Aerts, C., Thoul, A., Daszyńska, J., et al. 2003, Science, 300, 1926 [Google Scholar]
  7. Alcock, C., Allsman, R. A., Alves, D., et al. 1997, AJ, 114, 326 [NASA ADS] [CrossRef] [Google Scholar]
  8. Andrievsky, S. M., Kovtyukh, V. V., Korotin, S. A., Spite, M., & Spite, F. 2001, A&A, 367, 605 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  9. Ardeberg, A. 1980, A&AS, 42, 1 [NASA ADS] [Google Scholar]
  10. Ardeberg, A., & Maurice, E. 1977, A&AS, 30, 261 [NASA ADS] [Google Scholar]
  11. Ardeberg, A., Brunet, J. P., Maurice, E., & Prevot, L. 1972, A&AS, 6, 249 [NASA ADS] [Google Scholar]
  12. Asplund, M., Grevesse, N., Sauval, A. J., & Scott, P. 2009, ARA&A, 47, 481 [NASA ADS] [CrossRef] [Google Scholar]
  13. Azzopardi, M., & Vigneau, J. 1982, A&AS, 50, 291 [NASA ADS] [Google Scholar]
  14. Azzopardi, M., Vigneau, J., & Macquet, M. 1975, A&AS, 22, 285 [NASA ADS] [Google Scholar]
  15. Bailey, M., van Loon, J. T., Sarre, P. J., & Beckman, J. E. 2015, MNRAS, 454, 4013 [Google Scholar]
  16. Belczynski, K., Bulik, T., Fryer, C. L., et al. 2010, ApJ, 714, 1217 [NASA ADS] [CrossRef] [Google Scholar]
  17. Belczynski, K., Holz, D. E., Bulik, T., & O’Shaughnessy, R. 2016, Nature, 534, 512 [NASA ADS] [CrossRef] [Google Scholar]
  18. Bell, S. A., Hill, G., Hilditch, R. W., et al. 1991, MNRAS, 250, 119 [NASA ADS] [CrossRef] [Google Scholar]
  19. Berg, D. A., James, B. L., King, T., et al. 2022, ApJS, 261, 31 [NASA ADS] [CrossRef] [Google Scholar]
  20. Bestenlehner, J. M., Gräfener, G., Vink, J. S., et al. 2014, A&A, 570, A38 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  21. Björklund, R., Sundqvist, J. O., Puls, J., & Najarro, F. 2021, A&A, 648, A36 [EDP Sciences] [Google Scholar]
  22. Blair, W. P., Oliveira, C., LaMassa, S., et al. 2009, PASP, 121, 634 [NASA ADS] [CrossRef] [Google Scholar]
  23. Bohlin, R. C., Hubeny, I., & Rauch, T. 2020, AJ, 160, 21 [Google Scholar]
  24. Bonnarel, F., Fernique, P., Bienaymé, O., et al. 2000, A&AS, 143, 33 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  25. Bouchet, P., Lequeux, J., Maurice, E., Prevot, L., & Prevot-Burnichon, M. L. 1985, A&A, 149, 330 [NASA ADS] [Google Scholar]
  26. Bouret, J. C., Lanz, T., Hillier, D. J., et al. 2003, ApJ, 595, 1182 [NASA ADS] [CrossRef] [Google Scholar]
  27. Bouret, J. C., Lanz, T., Martins, F., et al. 2013, A&A, 555, A1 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  28. Bouret, J. C., Martins, F., Hillier, D. J., et al. 2021, A&A, 647, A134 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  29. Bowman, D. M. 2020, Front. Astron. Space Sci., 7, 70 [Google Scholar]
  30. Bowman, D. M., Burssens, S., Pedersen, M. G., et al. 2019, Nat. Astron., 3, 760 [Google Scholar]
  31. Brands, S. A., de Koter, A., Bestenlehner, J. M., et al. 2022, A&A, 663, A36 [Google Scholar]
  32. Brott, I., de Mink, S. E., Cantiello, M., et al. 2011, A&A, 530, A115 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  33. Burssens, S., Simón-Díaz, S., Bowman, D. M., et al. 2020, A&A, 639, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  34. Buscombe, W., & Foster, B. E. 1995, MK Spectral Classifications. Twelfth General Catalogue, Epoch 2000 Including UBV Photometry [Google Scholar]
  35. Cardelli, J. A., Clayton, G. C., & Mathis, J. S. 1989, ApJ, 345, 245 [Google Scholar]
  36. Carnall, A. C., Begley, R., McLeod, D. J., et al. 2023, MNRAS, 518, L45 [Google Scholar]
  37. Cartledge, S. I. B., Clayton, G. C., Gordon, K. D., et al. 2005, ApJ, 630, 355 [NASA ADS] [CrossRef] [Google Scholar]
  38. Castro, N., Oey, M. S., Fossati, L., & Langer, N. 2018, ApJ, 868, 57 [NASA ADS] [CrossRef] [Google Scholar]
  39. Chen, T. W., Smartt, S. J., Jerkstrand, A., et al. 2015, MNRAS, 452, 1567 [NASA ADS] [CrossRef] [Google Scholar]
  40. Cioni, M. R. L., Clementini, G., Girardi, L., et al. 2011, A&A, 527, A116 [CrossRef] [EDP Sciences] [Google Scholar]
  41. Conti, P. S., Garmany, C. D., & Massey, P. 1986, AJ, 92, 48 [NASA ADS] [CrossRef] [Google Scholar]
  42. Cox, N. L. J., Cordiner, M. A., Cami, J., et al. 2006, A&A, 447, 991 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  43. Cox, N. L. J., Cordiner, M. A., Ehrenfreund, P., et al. 2007, A&A, 470, 941 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  44. Crampton, D., & Greasley, J. 1982, PASP, 94, 31 [NASA ADS] [CrossRef] [Google Scholar]
  45. Crowther, P. A., & Smith, L. J. 1997, A&A, 320, 500 [NASA ADS] [Google Scholar]
  46. Crowther, P. A., & Walborn, N. R. 2011, MNRAS, 416, 1311 [Google Scholar]
  47. Crowther, P. A., Dessart, L., Hillier, D. J., Abbott, J. B., & Fullerton, A. W. 2002a, A&A, 392, 653 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  48. Crowther, P. A., Hillier, D. J., Evans, C. J., et al. 2002b, ApJ, 579, 774 [Google Scholar]
  49. Crowther, P. A., Lennon, D. J., & Walborn, N. R. 2006, A&A, 446, 279 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  50. Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2003, VizieR Online Data Catalog: II/246 [Google Scholar]
  51. Cutri, R. M., Skrutskie, M. F., van Dyk, S., et al. 2012, VizieR Online Data Catalog: II/281 [Google Scholar]
  52. Curti, M., D’Eugenio, F., Carniani, S., et al. 2023, MNRAS, 518, 425 [Google Scholar]
  53. David-Uraz, A., Neiner, C., Sikora, J., et al. 2019, MNRAS, 487, 304 [NASA ADS] [CrossRef] [Google Scholar]
  54. De Barros, S., Oesch, P. A., Labbé, I., et al. 2019, MNRAS, 489, 2355 [NASA ADS] [CrossRef] [Google Scholar]
  55. Dopita, M. A., Seitenzahl, I. R., Sutherland, R. S., et al. 2019, AJ, 157, 50 [Google Scholar]
  56. Doran, E. I., Crowther, P. A., de Koter, A., et al. 2013, A&A, 558, A134 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  57. Dufton, P. L., Ryans, R. S. I., Trundle, C., et al. 2005, A&A, 434, 1125 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  58. Dufton, P. L., Ryans, R. S. I., Simón-Díaz, S., Trundle, C., & Lennon, D. J. 2006, A&A, 451, 603 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  59. Dufton, P. L., Thompson, A., Crowther, P. A., et al. 2018, A&A, 615, A101 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  60. Dufton, P. L., Evans, C. J., Hunter, I., Lennon, D. J., & Schneider, F. R. N. 2019, A&A, 626, A50 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  61. Dupret, M. A., Thoul, A., Scuflaire, R., et al. 2004, A&A, 415, 251 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  62. Ehrenfreund, P., Cami, J., Jiménez-Vicente, J., et al. 2002, ApJ, 576, L117 [NASA ADS] [CrossRef] [Google Scholar]
  63. Ekström, S., Georgy, C., Eggenberger, P., et al. 2012, A&A, 537, A146 [Google Scholar]
  64. Erba, C., David-Uraz, A., Petit, V., et al. 2021, MNRAS, 506, 5373 [NASA ADS] [CrossRef] [Google Scholar]
  65. Evans, C. J., Crowther, P. A., Fullerton, A. W., & Hillier, D. J. 2004a, ApJ, 610, 1021 [NASA ADS] [CrossRef] [Google Scholar]
  66. Evans, C. J., Howarth, I. D., Irwin, M. J., Burnley, A. W., & Harries, T. J. 2004b, MNRAS, 353, 601 [NASA ADS] [CrossRef] [Google Scholar]
  67. Evans, C. J., Lennon, D. J., Trundle, C., Heap, S. R., & Lindler, D. J. 2004c, ApJ, 607, 451 [NASA ADS] [CrossRef] [Google Scholar]
  68. Evans, C. J., Lennon, D. J., Smartt, S. J., & Trundle, C. 2006, A&A, 456, 623 [CrossRef] [EDP Sciences] [Google Scholar]
  69. Evans, C. J., Bresolin, F., Urbaneja, M. A., et al. 2007, ApJ, 659, 1198 [Google Scholar]
  70. Evans, C. J., Taylor, W. D., Hénault-Brunet, V., et al. 2011, A&A, 530, A108 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  71. Evans, C. J., Kennedy, M. B., Dufton, P. L., et al. 2015a, A&A, 574, A13 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  72. Evans, C. J., van Loon, J. T., Hainich, R., & Bailey, M. 2015b, A&A, 584, A5 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  73. Fariña, C., Bosch, G. L., Morrell, N. I., Barbá, R. H., & Walborn, N. R. 2009, AJ, 138, 510 [CrossRef] [Google Scholar]
  74. Feitzinger, J. V., & Isserstedt, J. 1983, A&AS, 51, 505 [NASA ADS] [Google Scholar]
  75. Ferland, G. J., Chatzikos, M., Guzmán, F., et al. 2017, Rev. Mex. Astron. Astrofis., 53, 385 [NASA ADS] [Google Scholar]
  76. Firnstein, M., & Przybilla, N. 2012, A&A, 543, A80 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  77. Fitzgerald, M. P. 1970, A&A, 4, 234 [NASA ADS] [Google Scholar]
  78. Fitzpatrick, E. L. 1986, AJ, 92, 1068 [NASA ADS] [CrossRef] [Google Scholar]
  79. Fitzpatrick, E. L. 1988, ApJ, 335, 703 [NASA ADS] [CrossRef] [Google Scholar]
  80. Fitzpatrick, E. L. 1991, PASP, 103, 1123 [NASA ADS] [CrossRef] [Google Scholar]
  81. Fitzpatrick, E. L., Massa, D., Gordon, K. D., Bohlin, R., & Clayton, G. C. 2019, ApJ, 886, 108 [Google Scholar]
  82. Flores, B. L., & Hillier, D. J. 2021, MNRAS, 504, 311 [CrossRef] [Google Scholar]
  83. Foellmi, C., Moffat, A. F. J., & Guerrero, M. A. 2003, MNRAS, 338, 360 [NASA ADS] [CrossRef] [Google Scholar]
  84. Fullerton, A. W., Massa, D. L., & Prinja, R. K. 2006, ApJ, 637, 1025 [Google Scholar]
  85. Gal-Yam, A. 2012, Science, 337, 927 [Google Scholar]
  86. Garcia, M., Herrero, A., Najarro, F., Lennon, D. J., & Alejandro Urbaneja, M. 2014, ApJ, 788, 64 [NASA ADS] [CrossRef] [Google Scholar]
  87. Garcia, M., Herrero, A., Najarro, F., Camacho, I., & Lorenzo, M. 2019, MNRAS, 484, 422 [Google Scholar]
  88. Garland, R., Dufton, P. L., Evans, C. J., et al. 2017, A&A, 603, A91 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  89. Garmany, C. D., & Walborn, N. R. 1987, PASP, 99, 240 [NASA ADS] [CrossRef] [Google Scholar]
  90. Garmany, C. D., Conti, P. S., & Massey, P. 1987, AJ, 93, 1070 [NASA ADS] [CrossRef] [Google Scholar]
  91. Garmany, C. D., Massey, P., & Parker, J. W. 1994, AJ, 108, 1256 [NASA ADS] [CrossRef] [Google Scholar]
  92. Garnett, D. R. 1999, in New Views of the Magellanic Clouds, eds. Y. H. Chu, N. Suntzeff, J. Hesser, & D. Bohlender, 190, 266 [Google Scholar]
  93. Goldoni, P. 2011, Astron. Nachr., 332, 227 [NASA ADS] [CrossRef] [Google Scholar]
  94. Gordon, K. D., Clayton, G. C., Misselt, K. A., Landolt, A. U., & Wolff, M. J. 2003, ApJ, 594, 279 [NASA ADS] [CrossRef] [Google Scholar]
  95. Gordon, K. D., Cartledge, S., & Clayton, G. C. 2009, ApJ, 705, 1320 [NASA ADS] [CrossRef] [Google Scholar]
  96. Götberg, Y., de Mink, S. E., Groh, J. H., et al. 2018, A&A, 615, A78 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  97. Graczyk, D., Soszyński, I., Poleski, R., et al. 2011, Acta Astron., 61, 103 [Google Scholar]
  98. Graczyk, D., Pietrzyński, G., Thompson, I. B., et al. 2020, ApJ, 904, 13 [Google Scholar]
  99. Gräfener, G., Koesterke, L., & Hamann, W. R. 2002, A&A, 387, 244 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  100. Grin, N. J., Ramírez-Agudelo, O. H., de Koter, A., et al. 2017, A&A, 600, A82 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  101. Groh, J. H., Meynet, G., Ekström, S., & Georgy, C. 2014, A&A, 564, A30 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  102. Grunhut, J. H., Wade, G. A., Neiner, C., et al. 2017, MNRAS, 465, 2432 [NASA ADS] [CrossRef] [Google Scholar]
  103. Gvaramadze, V. V., Kniazev, A. Y., Maryeva, O. V., & Berdnikov, L. N. 2018, MNRAS, 474, 1412 [NASA ADS] [CrossRef] [Google Scholar]
  104. Haefner, R., Simon, K. P., & Fiedler, A. 1994, A&A, 288, L9 [NASA ADS] [Google Scholar]
  105. Hainich, R., Rühling, U., Todt, H., et al. 2014, A&A, 565, A27 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  106. Hainich, R., Pasemann, D., Todt, H., et al. 2015, A&A, 581, A21 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  107. Hamann, W. R., & Gräfener, G. 2003, A&A, 410, 993 [CrossRef] [EDP Sciences] [Google Scholar]
  108. Hamann, W. R., Feldmeier, A., & Oskinova, L. M. 2008, Clumping in hot-star winds: Proc. International Workshop held in Potsdam, Germany, 18-22 June 2007, eds. W.-R. Hamann, A. Feldmeier,& L. M. Oskinova [Google Scholar]
  109. Hawcroft, C., Sana, H., Mahy, L., et al. 2021, A&A, 655, A67 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  110. Hawcroft, C., Sana, H., Mahy, L., et al. 2023, A&A in press https://doi.org/10.1051/0004-6361/202245588 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  111. Heap, S. R., Lanz, T., & Hubeny, I. 2006, ApJ, 638, 409 [Google Scholar]
  112. Herrero, A., Kudritzki, R. P., Vilchez, J. M., et al. 1992, A&A, 261, 209 [NASA ADS] [Google Scholar]
  113. Heydari-Malayeri, M., & Hutsemekers, D. 1991, A&A, 243, 401 [NASA ADS] [Google Scholar]
  114. Hill, V. 1997, A&A, 324, 435 [Google Scholar]
  115. Hill, V. 1999, A&A, 345, 430 [NASA ADS] [Google Scholar]
  116. Hill, V., Andrievsky, S., & Spite, M. 1995, A&A, 293, 347 [NASA ADS] [Google Scholar]
  117. Hill, V., Barbuy, B., & Spite, M. 1997, A&A, 323, 461 [NASA ADS] [Google Scholar]
  118. Hillier, D. J. 2020, Galaxies, 8, 60 [NASA ADS] [CrossRef] [Google Scholar]
  119. Hillier, D. J., & Miller, D. L. 1998, ApJ, 496, 407 [NASA ADS] [CrossRef] [Google Scholar]
  120. Hosek, Jr., M. W., Kudritzki, R. P., Bresolin, F., et al. 2014, ApJ, 785, 151 [NASA ADS] [CrossRef] [Google Scholar]
  121. Howarth, I. D. 1983, MNRAS, 203, 301 [NASA ADS] [CrossRef] [Google Scholar]
  122. Hunter, I., Dufton, P. L., Ryans, R. S. I., et al. 2005, A&A, 436, 687 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  123. Hunter, I., Dufton, P. L., Smartt, S. J., et al. 2007, A&A, 466, 277 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  124. Hunter, I., Brott, I., Lennon, D. J., et al. 2008a, ApJ, 676, L29 [NASA ADS] [CrossRef] [Google Scholar]
  125. Hunter, I., Lennon, D. J., Dufton, P. L., et al. 2008b, A&A, 479, 541 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  126. Hunter, I., Brott, I., Langer, N., et al. 2009, A&A, 496, 841 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  127. Hutchings, J. B., Crampton, D., & Cowley, A. P. 1978, ApJ, 225, 548 [NASA ADS] [CrossRef] [Google Scholar]
  128. Isserstedt, J. 1975, A&AS, 19, 259 [NASA ADS] [Google Scholar]
  129. Isserstedt, J. 1979, A&AS, 38, 239 [NASA ADS] [Google Scholar]
  130. Isserstedt, J. 1982, A&AS, 50, 7 [NASA ADS] [Google Scholar]
  131. James, B. L., Berg, D. A., King, T., et al. 2022, ApJS, 262, 37 [NASA ADS] [CrossRef] [Google Scholar]
  132. Kaper, L., Henrichs, H. F., Nichols, J. S., et al. 1996, A&AS, 116, 257 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  133. Kaper, L., Henrichs, H. F., Nichols, J. S., & Telting, J. H. 1999, A&A, 344, 231 [NASA ADS] [Google Scholar]
  134. Kausch, W., Noll, S., Smette, A., et al. 2015, A&A, 576, A78 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  135. Kehrig, C., Vílchez, J. M., Pérez-Montero, E., et al. 2015, ApJ, 801, L28 [NASA ADS] [CrossRef] [Google Scholar]
  136. Koenigsberger, G., Morrell, N., Hillier, D. J., et al. 2014, AJ, 148, 62 [Google Scholar]
  137. Köhler, K., Langer, N., de Koter, A., et al. 2015, A&A, 573, A71 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  138. Korn, A. J., Becker, S. R., Gummersbach, C. A., & Wolf, B. 2000, A&A, 353, 655 [NASA ADS] [Google Scholar]
  139. Korn, A. J., Nieva, M. F., Daflon, S., & Cunha, K. 2005, ApJ, 633, 899 [NASA ADS] [CrossRef] [Google Scholar]
  140. Kurt, C. M., & Dufour, R. J. 1998, Rev. Mex. Astron. Astrofis. Conf. Ser., 7, 202 [NASA ADS] [Google Scholar]
  141. Krtička, J., & Kubát, J. 2017, A&A, 606, A31 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  142. Krtička, J., & Kubát, J. 2018, A&A, 612, A20 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  143. Kruckow, M. U., Tauris, T. M., Langer, N., et al. 2016, A&A, 596, A58 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  144. Kudritzki, R. P. 2002, ApJ, 577, 389 [NASA ADS] [CrossRef] [Google Scholar]
  145. Lamb, J. B., Oey, M. S., Graus, A. S., Adams, F. C., & Segura-Cox, D. M. 2013, ApJ, 763, 101 [NASA ADS] [CrossRef] [Google Scholar]
  146. Lamb, J. B., Oey, M. S., Segura-Cox, D. M., et al. 2016, ApJ, 817, 113 [NASA ADS] [CrossRef] [Google Scholar]
  147. Lamers, H. J. G. L. M., Cerruti-Sola, M., & Perinotto, M. 1987, ApJ, 314, 726 [NASA ADS] [CrossRef] [Google Scholar]
  148. Lamers, H. J. G. L. M., Zickgraf, F.-J., de Winter, D., Houziaux, L., & Zorec, J. 1998, A&A, 340, 117 [NASA ADS] [Google Scholar]
  149. Lanz, T., & Hubeny, I. 2003, ApJS, 146, 417 [NASA ADS] [CrossRef] [Google Scholar]
  150. Laplace, E., Götberg, Y., de Mink, S. E., Justham, S., & Farmer, R. 2020, A&A, 637, A6 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  151. Lebouteiller, V., Bernard-Salas, J., Brandl, B., et al. 2008, ApJ, 680, 398 [NASA ADS] [CrossRef] [Google Scholar]
  152. Lecoanet, D., Bowman, D. M., & Van Reeth, T. 2022, MNRAS, 512, L16 [NASA ADS] [CrossRef] [Google Scholar]
  153. Lennon, D. J. 1997, A&A, 317, 871 [NASA ADS] [Google Scholar]
  154. Lorenzo, M., Garcia, M., Najarro, F., et al. 2022, MNRAS, 516, 4164 [NASA ADS] [CrossRef] [Google Scholar]
  155. Maeder, A. 1987, A&A, 178, 159 [NASA ADS] [Google Scholar]
  156. Maeder, A., Przybilla, N., Nieva, M.-F., et al. 2014, A&A, 565, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  157. Magg, E., Bergemann, M., Serenelli, A., et al. 2022, A&A, 661, A140 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  158. Mahy, L., Sana, H., Abdul-Masih, M., et al. 2020, A&A, 634, A118 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  159. Maíz Apellániz, J., Evans, C. J., Barbá, R. H., et al. 2014, A&A, 564, A63 [CrossRef] [EDP Sciences] [Google Scholar]
  160. Maíz Apellániz, J., Barbá, R. H., Caballero, J. A., Bohlin, R. C., & Fariña, C. 2021, MNRAS, 501, 2487 [CrossRef] [Google Scholar]
  161. Mandel, I., & de Mink, S. E. 2016, MNRAS, 458, 2634 [NASA ADS] [CrossRef] [Google Scholar]
  162. Marcolino, W. L. F., Bouret, J. C., Rocha-Pinto, H. J., Bernini-Peron, M., & Vink, J. S. 2022, MNRAS, 511, 5104 [NASA ADS] [CrossRef] [Google Scholar]
  163. Martins, F., & Plez, B. 2006, A&A, 457, 637 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  164. Martins, F., Hillier, D. J., Bouret, J. C., et al. 2009, A&A, 495, 257 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  165. Massa, D., Fullerton, A. W., Nichols, J. S., et al. 1995, ApJ, 452, L53 [NASA ADS] [Google Scholar]
  166. Massa, D., Fullerton, A. W., Sonneborn, G., & Hutchings, J. B. 2003, ApJ, 586, 996 [NASA ADS] [CrossRef] [Google Scholar]
  167. Massey, P. 2002, ApJS, 141, 81 [NASA ADS] [CrossRef] [Google Scholar]
  168. Massey, P., Parker, J. W., & Garmany, C. D. 1989, AJ, 98, 1305 [Google Scholar]
  169. Massey, P., Lang, C. C., Degioia-Eastwood, K., & Garmany, C. D. 1995, ApJ, 438, 188 [NASA ADS] [CrossRef] [Google Scholar]
  170. Massey, P., Waterhouse, E., & DeGioia-Eastwood, K. 2000, AJ, 119, 2214 [NASA ADS] [CrossRef] [Google Scholar]
  171. Massey, P., Bresolin, F., Kudritzki, R. P., Puls, J., & Pauldrach, A. W. A. 2004, ApJ, 608, 1001 [Google Scholar]
  172. Massey, P., Puls, J., Pauldrach, A. W. A., et al. 2005, ApJ, 627, 477 [NASA ADS] [CrossRef] [Google Scholar]
  173. Massey, P., Olsen, K. A. G., Hodge, P. W., et al. 2007, AJ, 133, 2393 [Google Scholar]
  174. Massey, P., Zangari, A. M., Morrell, N. I., et al. 2009, ApJ, 692, 618 [Google Scholar]
  175. Massey, P., Neugent, K. F., & Morrell, N. 2017, ApJ, 837, 122 [NASA ADS] [CrossRef] [Google Scholar]
  176. Mathews, W. G. 1967, ApJ, 147, 965 [NASA ADS] [CrossRef] [Google Scholar]
  177. McEvoy, C. M., Dufton, P. L., Evans, C. J., et al. 2015, A&A, 575, A70 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  178. Mennickent, R. E., Cidale, L., Pietrzyński, G., Gieren, W., & Sabogal, B. 2006, A&A, 457, 949 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  179. Moehler, S., Modigliani, A., Freudling, W., et al. 2014, A&A, 568, A9 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  180. Mokiem, M. R., de Koter, A., Evans, C. J., et al. 2006, A&A, 456, 1131 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  181. Mokiem, M. R., de Koter, A., Evans, C. J., et al. 2007a, A&A, 465, 1003 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  182. Mokiem, M. R., de Koter, A., Vink, J. S., et al. 2007b, A&A, 473, 603 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  183. Morel, T., Castro, N., Fossati, L., et al. 2015, in New Windows on Massive Stars, eds. G. Meynet, C. Georgy, J. Groh, & P. Stee, 307, 342 [Google Scholar]
  184. Morrell, N. I., Niemela, V. S., Barbá, R. H., et al. 1999, in Revista Mexicana de Astronomia y Astrofisica Conference Series, eds. N. I. Morrell, V. S. Niemela, & R. H. Barbá, Rev. Mex. Astron. Astrofis. Conf. Ser., 8, 149 [NASA ADS] [Google Scholar]
  185. Negueruela, I., & Coe, M. J. 2002, A&A, 385, 517 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  186. Niemela, V., & Gamen, R. 2004, New A Rev., 48, 727 [NASA ADS] [CrossRef] [Google Scholar]
  187. Niemela, V. S. 2002, in Extragalactic Star Clusters, eds. D. P. Geisler, E. K. Grebel, & D. Minniti, 207, 202 [NASA ADS] [Google Scholar]
  188. Oey, M. S. 1996, ApJ, 467, 666 [NASA ADS] [CrossRef] [Google Scholar]
  189. Oey, M. S., & Kennicutt, R. C. 1997, MNRAS, 291, 827 [NASA ADS] [CrossRef] [Google Scholar]
  190. Oey, M. S., & Smedley, S. A. 1998, AJ, 116, 1263 [NASA ADS] [CrossRef] [Google Scholar]
  191. Oskinova, L. M., Hamann, W. R., & Feldmeier, A. 2007, A&A, 476, 1331 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  192. Ostrov, P. G., & Lapasset, E. 2003, MNRAS, 338, 141 [NASA ADS] [CrossRef] [Google Scholar]
  193. Palmerio, J. T., Vergani, S. D., Salvaterra, R., et al. 2019, A&A, 623, A26 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  194. Parker, J. W. 1993, AJ, 106, 560 [NASA ADS] [CrossRef] [Google Scholar]
  195. Parker, J. W., Garmany, C. D., Massey, P., & Walborn, N. R. 1992, AJ, 103, 1205 [NASA ADS] [CrossRef] [Google Scholar]
  196. Pauli, D., Oskinova, L. M., Hamann, W. R., et al. 2022, A&A, 659, A9 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  197. Pawlak, M., Graczyk, D., Soszyński, I., et al. 2013, Acta Astron., 63, 323 [NASA ADS] [Google Scholar]
  198. Paxton, B., Bildsten, L., Dotter, A., et al. 2011, ApJS, 192, 3 [Google Scholar]
  199. Paxton, B., Cantiello, M., Arras, P., et al. 2013, ApJS, 208, 4 [Google Scholar]
  200. Paxton, B., Marchant, P., Schwab, J., et al. 2015, ApJS, 220, 15 [Google Scholar]
  201. Paxton, B., Schwab, J., Bauer, E. B., et al. 2018, ApJS, 234, 34 [NASA ADS] [CrossRef] [Google Scholar]
  202. Paxton, B., Smolec, R., Schwab, J., et al. 2019, ApJS, 243, 10 [Google Scholar]
  203. Peimbert, A. 2003, ApJ, 584, 735 [NASA ADS] [CrossRef] [Google Scholar]
  204. Pellegrini, E. W., Oey, M. S., Winkler, P. F., et al. 2012, ApJ, 755, 40 [NASA ADS] [CrossRef] [Google Scholar]
  205. Penny, L. R., & Gies, D. R. 2009, ApJ, 700, 844 [Google Scholar]
  206. Penny, L. R., Sprague, A. J., Seago, G., & Gies, D. R. 2004, ApJ, 617, 1316 [NASA ADS] [CrossRef] [Google Scholar]
  207. Petrov, B., Vink, J. S., & Gräfener, G. 2016, MNRAS, 458, 1999 [NASA ADS] [CrossRef] [Google Scholar]
  208. Pietrzyński, G., Graczyk, D., Gallenne, A., et al. 2019, Nature, 567, 200 [Google Scholar]
  209. Prevot, M. L., Lequeux, J., Maurice, E., Prevot, L., & Rocca-Volmerange, B. 1984, A&A, 132, 389 [Google Scholar]
  210. Prinja, R. K., & Crowther, P. A. 1998, MNRAS, 300, 828 [NASA ADS] [CrossRef] [Google Scholar]
  211. Prinja, R. K., Barlow, M. J., & Howarth, I. D. 1990, ApJ, 361, 607 [NASA ADS] [CrossRef] [Google Scholar]
  212. Przybilla, N., Firnstein, M., Nieva, M. F., Meynet, G., & Maeder, A. 2010, A&A, 517, A38 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  213. Puls, J., Kudritzki, R. P., Herrero, A., et al. 1996, A&A, 305, 171 [Google Scholar]
  214. Puls, J., Springmann, U., & Lennon, M. 2000, A&AS, 141, 23 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  215. Puls, J., Urbaneja, M. A., Venero, R., et al. 2005, A&A, 435, 669 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  216. Puls, J., Vink, J. S., & Najarro, F. 2008, A&A Rv, 16, 209 [NASA ADS] [Google Scholar]
  217. Puls, J., Najarro, F., Sundqvist, J. O., & Sen, K. 2020, A&A, 642, A172 [EDP Sciences] [Google Scholar]
  218. Quimby, R. M., Kulkarni, S. R., Kasliwal, M. M., et al. 2011, Nature, 474, 487 [NASA ADS] [CrossRef] [Google Scholar]
  219. Ramírez-Agudelo, O. H., Sana, H., de Koter, A., et al. 2017, A&A, 600, A81 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  220. Ramachandran, V., Hainich, R., Hamann, W. R., et al. 2018a, A&A, 609, A7 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  221. Ramachandran, V., Hamann, W. R., Hainich, R., et al. 2018b, A&A, 615, A40 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  222. Ramachandran, V., Hamann, W. R., Oskinova, L. M., et al. 2019, A&A, 625, A104 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  223. Ramírez-Agudelo, O. H., Simón-Díaz, S., Sana, H., et al. 2013, A&A, 560, A29 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  224. Reynolds, A. P., Hilditch, R. W., Bell, S. A., & Hill, G. 1993, MNRAS, 261, 337 [NASA ADS] [CrossRef] [Google Scholar]
  225. Rickard, M. J., Hainich, R., Hamann, W. R., et al. 2022, A&A, 666, A189 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  226. Ricker, G. R., Winn, J. N., Vanderspek, R., et al. 2015, J. Astron. Telescopes Instrum. Syst., 1, 014003 [Google Scholar]
  227. Ritchie, B. W., Stroud, V. E., Evans, C. J., et al. 2012, A&A, 537, A29 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  228. Rivero González, J. G., Puls, J., Massey, P., & Najarro, F. 2012a, A&A, 543, A95 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  229. Rivero González, J. G., Puls, J., Najarro, F., & Brott, I. 2012b, A&A, 537, A79 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  230. Rolleston, W. R. J., Venn, K., Tolstoy, E., & Dufton, P. L. 2003, A&A, 400, 21 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  231. Roman-Duval, J., Proffitt, C. R., Taylor, J. M., et al. 2020, Res. Notes Am. Astron. Soc., 4, 205 [Google Scholar]
  232. Rousseau, J., Martin, N., Prévot, L., et al. 1978, A&AS, 31, 243 [NASA ADS] [Google Scholar]
  233. Sabín-Sanjulián, C., Simón-Díaz, S., Herrero, A., et al. 2014, A&A, 564, A39 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  234. Sabín-Sanjulián, C., Simón-Díaz, S., Herrero, A., et al. 2017, A&A, 601, A79 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  235. Salmon, S., Montalbán, J., Morel, T., et al. 2012, MNRAS, 422, 3460 [NASA ADS] [CrossRef] [Google Scholar]
  236. Salmon, S. J. A. J., Eggenberger, P., Montalbán, J., et al. 2022, A&A, 659, A142 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  237. Sana, H., de Mink, S. E., de Koter, A., et al. 2012, Science, 337, 444 [Google Scholar]
  238. Sana, H., de Koter, A., de Mink, S. E., et al. 2013, A&A, 550, A107 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  239. Sander, A. A. C., & Vink, J. S. 2020, MNRAS, 499, 873 [Google Scholar]
  240. Sander, A., Shenar, T., Hainich, R., et al. 2015, A&A, 577, A13 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  241. Sander, A. A. C., Hamann, W. R., Todt, H., Hainich, R., & Shenar, T. 2017, A&A, 603, A86 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  242. Sander, A. A. C., Vink, J. S., & Hamann, W. R. 2020, MNRAS, 491, 4406 [Google Scholar]
  243. Sanduleak, N. 1969, AJ, 74, 877 [NASA ADS] [CrossRef] [Google Scholar]
  244. Santolaya-Rey, A. E., Puls, J., & Herrero, A. 1997, A&A, 323, 488 [NASA ADS] [Google Scholar]
  245. Schmidt-Kaler, T., Gochermann, J., Oestreicher, M. O., et al. 1999, MNRAS, 306, 279 [NASA ADS] [CrossRef] [Google Scholar]
  246. Schneider, F. R. N., Langer, N., de Koter, A., et al. 2014, A&A, 570, A66 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  247. Schnurr, O., Moffat, A. F. J., St-Louis, N., Morrell, N. I., & Guerrero, M. A. 2008, MNRAS, 389, 806 [NASA ADS] [CrossRef] [Google Scholar]
  248. Seaton, M. J. 1979, MNRAS, 187, 73 [Google Scholar]
  249. Selman, F., Melnick, J., Bosch, G., & Terlevich, R. 1999, A&A, 341, 98 [NASA ADS] [Google Scholar]
  250. Shenar, T., Hainich, R., Todt, H., et al. 2016, A&A, 591, A22 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  251. Shenar, T., Hainich, R., Todt, H., et al. 2018, A&A, 616, A103 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  252. Shenar, T., Sablowski, D. P., Hainich, R., et al. 2019, A&A, 627, A151 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  253. Shenar, T., Sana, H., Mahy, L., et al. 2022, A&A, 665, A148 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  254. Shirazi, M., & Brinchmann, J. 2012, MNRAS, 421, 1043 [NASA ADS] [CrossRef] [Google Scholar]
  255. Simón-Díaz, S. 2020, Reviews in Frontiers of Modern Astrophysics (FromSpace Debris to Cosmology: From Space Debris to Cosmology), 155 [CrossRef] [Google Scholar]
  256. Simón-Díaz, S., & Herrero, A. 2014, A&A, 562, A135 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  257. Sirressi, M., Adamo, A., Hayes, M., et al. 2022, AJ, 164, 208 [NASA ADS] [CrossRef] [Google Scholar]
  258. Smette, A., Sana, H., Noll, S., et al. 2015, A&A, 576, A77 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  259. Smith, L. F., Shara, M. M., & Moffat, A. F. J. 1990, ApJ, 348, 471 [NASA ADS] [CrossRef] [Google Scholar]
  260. Smith, L. F., Shara, M. M., & Moffat, A. F. J. 1996, MNRAS, 281, 163 [NASA ADS] [CrossRef] [Google Scholar]
  261. Smith Neubig, M. M., & Bruhweiler, F. C. 1997, AJ, 114, 1951 [CrossRef] [Google Scholar]
  262. Spitzer, L. 1978, Physical Processes in the Interstellar Medium (New York: Wiley- Interscience) [Google Scholar]
  263. Steidel, C. C., Strom, A. L., Pettini, M., et al. 2016, ApJ, 826, 159 [NASA ADS] [CrossRef] [Google Scholar]
  264. Strom, A. L., Rudie, G. C., Steidel, C. C., & Trainor, R. F. 2022, ApJ, 925, 116 [NASA ADS] [CrossRef] [Google Scholar]
  265. Sundqvist, J. O., Owocki, S. P., & Puls, J. 2018, A&A, 611, A17 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  266. Sundqvist, J. O., Björklund, R., Puls, J., & Najarro, F. 2019, A&A, 632, A126 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  267. Tautvaišienė, G., Geisler, D., Wallerstein, G., et al. 2007, AJ, 134, 2318 [CrossRef] [Google Scholar]
  268. Tchernyshyov, K., Meixner, M., Seale, J., et al. 2015, ApJ, 811, 78 [NASA ADS] [CrossRef] [Google Scholar]
  269. Testor, G., & Niemela, V. 1998, A&AS, 130, 527 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  270. Toribio San Cipriano, L., Domínguez-Guzmán, G., Esteban, C., et al. 2017, MNRAS, 467, 3759 [NASA ADS] [CrossRef] [Google Scholar]
  271. Tramper, F., Sana, H., de Koter, A., Kaper, L., & Ramírez-Agudelo, O. H. 2014, A&A, 572, A36 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  272. Trundle, C., & Lennon, D. J. 2005, A&A, 434, 677 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  273. Trundle, C., Lennon, D. J., Puls, J., & Dufton, P. L. 2004, A&A, 417, 217 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  274. Trundle, C., Dufton, P. L., Hunter, I., et al. 2007, A&A, 471, 625 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  275. Tsamis, Y. G., Barlow, M. J., Liu, X. W., Danziger, I. J., & Storey, P. J. 2003, MNRAS, 338, 687 [NASA ADS] [CrossRef] [Google Scholar]
  276. ud-Doula, A., Owocki, S. P., Townsend, R. H. D., et al. 2008, MNRAS, 385, 97 [NASA ADS] [CrossRef] [Google Scholar]
  277. Ulaczyk, K., Szymański, M. K., Udalski, A., et al. 2013, Acta Astron., 63, 159 [NASA ADS] [Google Scholar]
  278. Urbaneja, M. A., Kudritzki, R. P., Gieren, W., et al. 2017, AJ, 154, 102 [NASA ADS] [CrossRef] [Google Scholar]
  279. van Loon, J. T., Bailey, M., Tatton, B. L., et al. 2013, A&A, 550, A108 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  280. Venn, K. A. 1999, ApJ, 518, 405 [Google Scholar]
  281. Vernet, J., Dekker, H., D’Odorico, S., et al. 2011, A&A, 536, A105 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  282. Verro, K., Trager, S. C., Peletier, R. F., et al. 2022, A&A, 660, A34 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  283. Vink, J. S., de Koter, A., & Lamers, H. J. G. L. M. 1999, A&A, 350, 181 [Google Scholar]
  284. Vink, J. S., de Koter, A., & Lamers, H. J. G. L. M. 2000, A&A, 362, 295 [Google Scholar]
  285. Vink, J. S., de Koter, A., & Lamers, H. J. G. L. M. 2001, A&A, 369, 574 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  286. Voges, E. S., Oey, M. S., Walterbos, R. A. M., & Wilkinson, T. M. 2008, AJ, 135, 1291 [NASA ADS] [CrossRef] [Google Scholar]
  287. Vreeswijk, P. M., Ellison, S. L., Ledoux, C., et al. 2004, A&A, 419, 927 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  288. Wade, G. A., Neiner, C., Alecian, E., et al. 2016, MNRAS, 456, 2 [Google Scholar]
  289. Walborn, N. R. 1977, ApJ, 215, 53 [NASA ADS] [CrossRef] [Google Scholar]
  290. Walborn, N. R. 1983, ApJ, 265, 716 [NASA ADS] [CrossRef] [Google Scholar]
  291. Walborn, N. R., Lennon, D. J., Haser, S. M., Kudritzki, R.-P., & Voels, S. A. 1995, PASP, 107, 104 [NASA ADS] [CrossRef] [Google Scholar]
  292. Walborn, N. R., Lennon, D. J., Heap, S. R., et al. 2000, PASP, 112, 1243 [Google Scholar]
  293. Walborn, N. R., Fullerton, A. W., Crowther, P. A., et al. 2002a, ApJS, 141, 443 [NASA ADS] [CrossRef] [Google Scholar]
  294. Walborn, N. R., Howarth, I. D., Lennon, D. J., et al. 2002b, AJ, 123, 2754 [NASA ADS] [CrossRef] [Google Scholar]
  295. Walborn, N. R., Morrell, N. I., Howarth, I. D., et al. 2004, ApJ, 608, 1028 [NASA ADS] [CrossRef] [Google Scholar]
  296. Walborn, N. R., Howarth, I. D., Evans, C. J., et al. 2010, AJ, 139, 1283 [NASA ADS] [CrossRef] [Google Scholar]
  297. Walborn, N. R., Sana, H., Simón-Díaz, S., et al. 2014, A&A, 564, A40 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  298. Weaver, R., McCray, R., Castor, J., Shapiro, P., & Moore, R. 1977, ApJ, 218, 377 [Google Scholar]
  299. Welty, D. E., Federman, S. R., Gredel, R., Thorburn, J. A., & Lambert, D. L. 2006, ApJS, 165, 138 [Google Scholar]
  300. Yoon, S. C., & Langer, N. 2005, A&A, 443, 643 [NASA ADS] [CrossRef] [EDP Sciences] [Google Scholar]
  301. Yoon, S.-C., Dessart, L., & Clocchiatti, A. 2017, ApJ, 840, 10 [NASA ADS] [CrossRef] [Google Scholar]
  302. Zastrow, J., Oey, M. S., & Pellegrini, E. W. 2013, ApJ, 769, 94 [NASA ADS] [CrossRef] [Google Scholar]
  303. Zaritsky, D., Harris, J., Thompson, I. B., Grebel, E. K., & Massey, P. 2002, AJ, 123, 855 [Google Scholar]
  304. Zaritsky, D., Harris, J., Thompson, I. B., & Grebel, E. K. 2004, AJ, 128, 1606 [Google Scholar]
  305. Zasche, P., Wolf, M., & Vraštil, J. 2017, MNRAS, 472, 2241 [NASA ADS] [CrossRef] [Google Scholar]
  306. Zickgraf, F. J., Kovacs, J., Wolf, B., et al. 1996, A&A, 309, 505 [Google Scholar]

Appendix A: The XShootU collaboration

A.1. Organisation (WG 1)

XShootU is a project with a wide-scale community approach. The collaboration is organised into 14 WGs that anyone can join: (1) Organisation, (2) Data Reduction and Calibration, (3) Stellar Atmospheres and Benchmarking, (4) Wind Structure, (5) Stellar Evolution Modelling, (6) Stellar Libraries, (7) Population Synthesis, (8) Interstellar Medium, (9) Massive Star Feedback, (10) Stripped Stars, (11), Auxiliary/New Data, (12) Pulsations, (13) Magnetic Fields, and (14) Unusual Objects.

A.2. Data Reduction and Calibration (WG 2)

The detailed description of the UVB and VIS higher level data products will be associated with DR1 (Sana et al., in prep., XShootU II). The data reduction of the NIR spectra will become part of DR2.

A.3. Stellar Atmospheres and Benchmarking (WG 3)

WG 3 was set up to foster discussions and cooperation between researchers performing the spectroscopic analysis of the XShootU datasets. Part of this work is described in the main part of the paper. Several sub-groups were organised by WG 3 members to address specific science questions and topics of interest. This ranges from the determination of the full photospheric and wind properties of the sample at large, to very specific issues such as studying the effects of rotational mixing on stars with different masses/ages through abundance studies. The three main codes that are employed for the spectroscopic analysis are the NLTE expanding wind codes CMFGEN, FASTWIND, and PoWR (Hillier & Miller 1998; Puls et al. 2020; Gräfener et al. 2002), and the plane parallel NLTE code TLUSTY for stars without strong winds.

A.4. Wind Structure (WG 4)

Winds of massive, hot stars are time-dependent and highly structured on small and large spatial scales (see the overviews in Puls et al. 2008; Hamann et al. 2008). It has been shown that properties of structured winds have an enormous impact on empirical estimates of mass-loss rates, and thus on our understanding of stellar evolution. However, the question to which extent Ṁ is influenced by wind structures remains. WG 4 studies the clumping dependence on spectral and luminosity classes at different Z. Spectral modelling is performed using state-of-the-art model atmosphere codes (see WG 3), which can treat clumping properties with different levels of sophistication. Wind clumping is usually treated in the ‘micro-clumping’ (i.e. optically thin clumps at all frequencies) or ‘macro-clumping’ (i.e. arbitrary optical thickness of clumps) approximations, still assuming a void inter-clump medium and a smooth velocity field (e.g. CMFGEN, PoWR, and METUJE; Krtička & Kubát 2017, 2018).

A.5. Stellar Evolution Modelling (WG 5)

The objectives of the stellar evolution group fall into two categories. The first one is to provide up to date sets of stellar evolution models covering not only the parameter space of initial conditions but also that of uncertain physical prescriptions such as overshooting. Following the work of Schneider et al. (2014), we also provide open source tools7 for performing Bayesian inference of observed systems and inferring their initial properties given a set of simulations. Even though the XShootU sample was designed to exclude known binary stars, it is possible that such a sample contains large numbers of binary products. Therefore, extended grids of binary evolution will also be constructed and incorporated into the Bayesian inference framework under construction. Our simulations will be coded using the MESA software instrument (Paxton et al. 2011, 2013, 2015, 2018, 2019), but our Bayesian tools are designed to easily incorporate simulations from other stellar evolution simulations, allowing us to establish systematic biases on our theoretical models. The second objective of the WG will be the development of tailored evolutionary models for specific objects of interest found within the XShootU sample.

A.6. Stellar Libraries (WG 6)

Stellar spectral libraries are one of the main ingredients of stellar population synthesis models, which in turn are powerful tools in the study of fundamental properties of unresolved stellar systems. One of the major deficiencies of the empirical libraries available today is the coverage of hot and young stars at low Z (Hill et al. 2022). A variety of empirical stellar spectral libraries, collected with different primary goals, are publicly available (see Table 1 in Verro et al. 2021 for a recent summary). In terms of spectral resolving power, only the X-shooter Spectral Library (XSL) and ELODIE are comparable to the XShootU dataset. ULLYSES + XShootU is by far the most complete, highest-S/N, and highest resolution library of hot, massive stars with the broadest spectral coverage. While numerous libraries for low-mass stars are available, these libraries are incomplete at high masses. To illustrate this point, we compared the ULLYSES + XShootU target sample with the XSL library (Verro et al. 2022), which was designed for stellar population synthesis. In Fig. A.1 we show the HR diagram coverage of the XSL library, which shows the lack of massive OB stars at any Z. ULLYSES + XShootU perfectly complement the missing parameter space of the XSL library. In combination, the two libraries allow self-consistent population synthesis models (see WG 7) of systems hosting both young and old stars.

thumbnail Fig. A.1.

Comparison of the ULLYSES (blue: LMC; green: SMC) and XSL libraries. XSL stars (filled circles) are separated by metallicity: [Fe/H] >  − 0.5 (black) and [Fe/H] < − 0.5 (dark red). Temperatures, gravities, and metallicities of the ULLYSES stars in this figure were obtained from the literature. Solid lines are Geneva isochrones for Z⊙ and various ages (Ekström et al. 2012).

A.7. Population Synthesis (WG 7)

Population synthesis models are called semi-empirical if the stellar evolution tracks are theoretical but the individual stellar spectra are observed, and fully theoretical when both are calculated. XShootU observations will enable improvements of semi-empirical models because they will provide the most complete spectral library to date. In addition, the XShootU observations will guide new generations of atmosphere and evolution models, which will help improve fully theoretical population synthesis models. The new empirical library of LMC + SMC UV-to-NIR massive-star spectra built by the ULLYSES and XShootU projects will replace the Starburst99 LMC + SMC library, greatly improving the realism of population synthesis predictions. Once the library is implemented in Starburst99, corresponding Cloudy (Ferland et al. 2017) photoionisation models will be computed in order to account for the contribution of the ionised gas and dust to the integrated light of young OB star populations. The models, which will be publicly available, will be used to interpret existing and future observations of star-forming and starburst galaxies with similar metallicities, in particular those from CLASSY (Berg et al. 2022; James et al. 2022) high-resolution (R ∼ 15 000) Far-UV spectral database of 45 nearby (0.002 < z < 0.182) star-forming galaxies, the majority of which have LMC- or SMC-like metallicities.

A.8. Interstellar Medium (WG 8)

The intervening ISM along the line of sight to each target star will impact the observed UV and optical spectra. Cool atomic and molecular gas produces absorption features superimposed on stellar photospheric and wind lines. Furthermore, dust in the Milky Way and the host galaxy extinguishes and reddens the stellar spectra. These effects must be accounted for to reliably analyse stellar features, but also enable ISM science that is interesting in its own right. Due to the nature of the UV-based selection process of the target sample and the fact that most OB stars in the Magellanic Clouds experience low extinctions (E(B − V) < 0.20 mag), strong ISM signatures are not expected in the optical and NIR X-shooter spectra. Nevertheless, clear signatures are expected in strong atomic lines such as the Na Iλλ5889.951,5895.924 and Ca IIλλ3933.663,3968.468 doublets. For the targets with the strongest extinctions, detection are expected for some molecular lines, such as CH λ4300.313 and CH+ λ4232.548, and strong diffuse interstellar bands, such as those at 4428 Å, 5780.5 Å, 5797.1 Å, and 6614 Å. These lines have been well studied for Galactic targets but less so for objects in the Magellanic Clouds (though see e.g. Ehrenfreund et al. 2002; Welty et al. 2006; Cox et al. 2006, 2007; van Loon et al. 2013; Bailey et al. 2015). The coverage of the significantly less studied wavelength region between 3200 − 3800 Å could provide detections of some lines such as Ti Iλ3729.8069 and Ti IIλ3383.759. In addition, the flux calibration of the X-shooter spectra will provide insights into the Magellanic Cloud extinction laws (Maíz Apellániz et al. 2014) and possibly detect the recently discovered very broad absorption band centred on 7700 Å (Maíz Apellániz et al. 2021).

A.9. Massive Star Formation and Feedback (WG 9)

The energy output from young OB stars creates expanding H II regions that are over-pressurised with respect to their surroundings by photoionisation, radiation pressure on dust grains, and shock heating by stellar winds (Mathews 1967; Weaver et al. 1977; Spitzer 1978). Key stellar properties for the structure and dynamics of these regions are the flux in far-UV and extreme-UV photons and the stellar wind mass-loss rate and terminal velocity. The density distribution, chemistry, and equilibrium temperature of the photoionised gas also play important roles (Ferland et al. 2017). Because the LMC and SMC have substantially lower metallicities than the Galaxy, their H II regions are ideal test cases for the impacts of stellar ionising flux (Voges et al. 2008; Oey & Kennicutt 1997) and winds (e.g. Oey 1996; Oey & Smedley 1998) in low-metallicity environments typical of high-redshift galaxies. The XShootU dataset in the Magellanic Clouds represents a significant opportunity to advance the state of the art in our understanding of how stars shape their environment. WG9 has identified a sample of relatively isolated XShootU stars that reside inside H II regions identified in the MCELS survey (Pellegrini et al. 2012). These single-star H II regions are valuable tools for calibrating stellar atmosphere models (like those produced by WG 3) by comparing observed nebular emission lines to photoionisation models adopting those model atmospheres as the ionising sources (Zastrow et al. 2013). Combining stellar properties inferred from the XShootU spectra with the observed structure and emission of the ionised gas in these regions will clarify the role stars play in shaping their environment.

A.10. Stripped Stars (WG 10)

Stars stripped of their hydrogen-rich envelopes via mass transfer or common envelope ejection in binary systems are thought to be the fate of a third of all massive stars (Sana et al. 2012). In fact, in a continuously star-forming population, a few percent of all massive stars should be accompanied by a stripped companion. Binary-stripped stars are the exposed helium cores of their progenitors. During the long-lived phase of core helium burning these stars are small (R⋆ ≲ 1 R⊙), hot (Teff ∼ 50 − 100 kK), compact (log g ∼ 5), helium-rich (Ysurf ∼ 0.5 − 1), and hydrogen-poor (Xsurf ∼ 0 − 0.5), but span a range of luminosities (1 − 106 L⊙) and masses (∼0.5 − 10 M⊙) (Götberg et al. 2018). After core-helium exhaustion, they can expand to giant sizes, depending on the fraction of hydrogen they retain, and appear as helium-enriched blue, yellow, or red supergiants (Yoon et al. 2017; Laplace et al. 2020). ULLYSES offers an excellent sample to search for hot binary-stripped stars that are sufficiently massive to explode as SNe. Especially with the effort to exclude apparent binaries from the target sample, the likelihood of having a few hidden binary-stripped companions increases, as they are not expected to significantly contribute to the optical flux or to cause large radial velocities. While UV spectra will be crucial for this analysis, the optical spectra are needed for stellar characterisation of both the binary-stripped star and the accretor star. We expect that discovering binary-stripped star + OB star systems in the ULLYSES sample will help constrain uncertainties, such as how common and efficient mass transfer is, and exactly how the two stars are affected by such mass transfer.

A.11. Auxiliary/New Data (WG 11)

There are multiple ways in which auxiliary data can help increase the value of the ULLYSES/XShootU library, including different wavelength regimes (e.g. X-rays) and observing strategies. For example, the intermediate spectral resolving power of the X-shooter data only partially resolves spectral lines, especially for stars with low projected rotational velocities (v sin i), and higher resolution spectra are needed for a detailed spectroscopic analysis. WG 11 will provide high-resolution optical spectroscopy of a selection of slow rotating ULLYSES targets. Targets were selected on the basis of showing narrow, resolved C III 1176 Å lines in their UV spectra or having low v sin i reported in the literature. Between December 2021 and December 2022, we obtained high resolution (R ∼ 40 000) and high S/N (∼200 in the 4100 Å region) optical spectra with the 6.5m Magellan Clay telescope (+MIKE spectrograph) for 48 ULLYSES targets. Basic reductions confirmed the high quality of the data, and final reductions are underway. The Magellan/MIKE data will be made available to the massive stars community. WG11 also plans to organise archival searches aimed to identify data that will enhance the ULLYSES/XShootU library:

  • High spectral resolution optical spectra of ULLYSES targets, for example obtained with the UVES, FLAMES, FEROS, and HARPS instruments.

  • UV spectra obtained with IUE, FUSE, and HST.

A.12. Pulsations (WG 12)

The study of spectroscopic variability in XShootU targets is naturally supported by the analysis of time-series photometry from the NASA Transiting Exoplanet Survey Satellite (TESS) mission (Ricker et al. 2015). The LMC is within the southern continuous viewing zone of the TESS mission, such that all XShootU stars in the LMC have continuous light curves spanning 1 yr. TESS light curves have been used to demonstrate that heat-driven pulsations and the signatures of rotational modulation are common in early-type main-sequence stars in the Galaxy (David-Uraz et al. 2019; Burssens et al. 2020). Whereas, the brightest massive stars in the Magellanic Clouds (i.e. OB supergiants) have light curves dominated by stochastic low-frequency variability (Bowman et al. 2019). Interestingly, in low-metallicity environments such as the Magellanic Clouds, the driving of heat-driven pulsations is less efficient (Salmon et al. 2012). For massive stars with identified pulsation modes, forward asteroseismic modelling is able to provide constraints on their (core) masses, ages, interior rotation profiles, and mixing properties (e.g. Aerts et al. 2003; Dupret et al. 2004; Salmon et al. 2022), but also probe the strength and geometry of an interior magnetic field (Lecoanet et al. 2022). A key goal of WG12 is to extract reliable TESS light curves, provide complementary constraints on the variability amplitudes and timescales of each XShootU target, and ultimately perform forward asteroseismic modelling (see Bowman 2020 for a review).

A.13. Magnetic Fields (WG13)

WG13 seeks to leverage the UV and optical spectral libraries provided by the ULLYSES and XShootU programmes to shed light on how magnetism affects massive star structure and evolution. In magnetic massive stars, which make up ∼7% of the total OB star population (e.g. Morel et al. 2015; Wade et al. 2016; Grunhut et al. 2017), stable, nearly dipolar surface magnetic fields warp the stellar wind into a structurally complex magnetosphere, trapping the wind near the stellar surface, thus significantly altering the circumstellar environment. This magnetic wind confinement has been shown to reduce stellar mass-loss rate compared to non-magnetic stars of similar spectral type (ud-Doula et al. 2008). It should be noted that atypical features in P-Cygni profiles are not necessarily the result of a magnetic fields, as variability in the absorption components of wind-sensitive UV resonance lines have also been observed in non-magnetic massive stars (e.g. Massa et al. 1995; Kaper et al. 1996, 1999). The Erba et al. (2021) synthetic line profiles can be adapted for direct comparison with observations, and used in conjunction with the XShootU spectral libraries to identify new magnetic candidates.

A.14. Unusual Objects (WG14)

Finally, the data offer many opportunities to study key objects. For example, combining the radial velocities from XShootU with the exquisite proper motions from Gaia DR3, we can reconstruct the space motions of ULLYSES targets, which can be used to identify runaway and walkaway stars. Moreover, Gaia DR3 light-curves can be used in conjunction with radial velocities of known eclipsing binaries to estimate fundamental stellar parameters, providing unique tests of stellar evolution.

Appendix B: ULLYSES target parameters

In Table B.1 we present optical and NIR (JHK) photometry of the ULLYSES targets, as well as spectroscopic and multiplicity information. Table B.2 provides an extensive literature search for the stellar and wind parameters as they were known prior to the start of ULLYSES. Obviously, these data (see the excerpt of Table B.2; the full table is available at the CDS) are rather heterogeneous and they will quickly become outdated as new spectral analyses are underway.

Table B.1.

Photometric, spectroscopic, and multiplicity of ULLYSES/XShootU targets, ordered by RA (Magellan/MIKE targets are indicated in the final column).

Table B.2.

Pre-ULLYSES target parameters compiled by the ULLYSES team at STScI. Excerpt only. The full table is available at the CDS.

All Tables

Table 1.

Baseline LMC metal abundances (X/H by number) with respect to Z⊙ from Magg et al. (2022, MBS22).

In the text
Table 2.

Baseline SMC metal abundances (X/H by number) adopted with respect to Z⊙ from Magg et al. (2022, MBS22).

In the text
Table B.1.

Photometric, spectroscopic, and multiplicity of ULLYSES/XShootU targets, ordered by RA (Magellan/MIKE targets are indicated in the final column).

In the text
Table B.2.

Pre-ULLYSES target parameters compiled by the ULLYSES team at STScI. Excerpt only. The full table is available at the CDS.

In the text

All Figures

thumbnail Fig. 1.

Positions of the ULLYSES/XShootU sources in the LMC (left) and SMC (right). Yellow dots are O-type stars, red diamonds are B-type stars, and blue squares are WR and WR-like Of/WR ‘slash’ stars. We note that the two images have different spatial scales. This figure was made with the Aladin Sky Atlas (Bonnarel et al. 2000); the background consists of DSS2 colour images.
In the text
thumbnail Fig. 2.

Distribution of the spectral types in ULLYSES. For the ten known binaries in the sample, only the primary component is accounted for. The category labelled ‘W’ includes WR and WR-like ‘slash’ stars.
In the text
thumbnail Fig. 3.

HR diagrams of the ULLYSES SMC and LMC targets. Stellar parameters are based on contemporary literature from Table B.2 (filled symbols) or spectral type calibrations (open symbols). For known binary and multiple systems, only primaries are indicated. Calibrations used are Doran et al. (2013) for O-type stars in both galaxies, Dufton et al. (2019), Trundle et al. (2004), and Trundle & Lennon (2005) for SMC B-type stars, and Dufton et al. (2018), Garland et al. (2017), McEvoy et al. (2015), and Urbaneja et al. (2017) for LMC B-type stars. Evolutionary tracks (solid lines) and isochrones (dotted lines) for non-rotating massive stars at 0.5 Z⊙ and 0.2 Z⊙ are from Brott et al. (2011), supplemented by tracks for very massive stars in the LMC from Köhler et al. (2015).
In the text
thumbnail Fig. 4.

Main-sequence MESA stellar evolution models of a rapidly rotating (v sin i = 550 km s−1) 50 M⊙ star for a range of metallicities. The Galactic and LMC models show traditional redward evolution, while the SMC and even lower-Z (one-tenth solar) models start to show blueward chemical homogeneous evolution. The models employ Vink et al. (2000, 2001) mass-loss rates and assume a moderate amount of core overshooting, with a value of αov of 0.335, similar to the Brott et al. (2011) models.
In the text
thumbnail Fig. 5.

Effect of spectral resolution and rotational velocity on three sets of optical lines classically used to determine C, N, and O abundances. In each panel, the initial CMFGEN model has Teff = 31 000 K, log g = 3.6 and 0.2 Z⊙. The model is degraded to a resolution of either 6000 (typical for our X-shooter UVB spectra) or 80 000 and further convolved with three rotational velocities (20, 100, and 300 km s−1). No additional macro-turbulent broadening is considered. Also plotted is the UV C III 1176 line. Here the spectral resolution is that of the STIS E140M grating (R ∼ 45 000).
In the text
thumbnail Fig. 6.

Reduced X-shooter spectra for a range of spectral types of single-star supergiants (top) and dwarfs (bottom). For illustration purposes, the flux of each spectrum was divided by its mean value and an arbitrary offset was added. The grey regions correspond to the UVB-VIS common wavelength coverage (∼ 5500 Å), a gap due to bad pixel masking (∼ 6360 Å), and telluric absorption. Minor manual treatment to remove strong cosmic rays was performed.
In the text
thumbnail Fig. 7.

X-shooter/UBV spectrum of Sk −67° 167 (O4 Inf+) in the LMC, including zoomed-in views of key spectroscopic diagnostics.
In the text
thumbnail Fig. 8.

Comparison between the observed HST (top) and X-shooter spectrum (black line) of selected lines of C, N, and O for AzV 327 (O9.5 II-Ibw) with two models (coloured lines). The light blue line is for solar-scaled abundances (factor of 1/5), while the red model has the following scaling: C abundance decreased by a factor of 3.8, N abundance increased by a factor of 4.5, and O abundance decreased by a factor of 1.6. The models were computed with the NLTE CMFGEN (Hillier & Miller 1998) atmosphere code.
In the text
thumbnail Fig. 9.

UV (top) and optical (bottom) spectrum of the LMC star Sk-67 167 (O4 Inf+). The UV spectrum consists of STIS E140M observations taken as part of the ULLYSES project and archival FUSE data. The optical spectrum was obtained with X-shooter. A selection of diagnostics for stellar and wind parameters are highlighted. Note that these diagnostics can vary with spectral type. The ticks at the bottom of the UV spectra mark the position of interstellar lines.
In the text
thumbnail Fig. 10.

UV (top, middle) and optical (bottom) spectrum of the SMC star Sk191 (B1.5 Ia). The UV spectrum consists of FUSE, STIS E140M, and STIS E230M observations compiled as part of the ULLYSES project. The optical spectrum was obtained with X-shooter. Similarly to Fig. 9, interstellar transitions and a selection of stellar and wind diagnostics are highlighted. In addition, metallic lines that can be used to measure abundances are marked in purple.
In the text
thumbnail Fig. 11.

O giant (AV186, O8.5III). The best fitting model is for T = 33 kK and log g = 3.4. Note that while a ten times higher mass star with log g = 4.4 would be indistinguishable in the UV, it would completely fail to reproduce the optical Balmer wings. The model was computed with the PoWR (Sander et al. 2017) NLTE code.
In the text
thumbnail Fig. A.1.

Comparison of the ULLYSES (blue: LMC; green: SMC) and XSL libraries. XSL stars (filled circles) are separated by metallicity: [Fe/H] >  − 0.5 (black) and [Fe/H] < − 0.5 (dark red). Temperatures, gravities, and metallicities of the ULLYSES stars in this figure were obtained from the literature. Solid lines are Geneva isochrones for Z⊙ and various ages (Ekström et al. 2012).
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.