GHGonzalo Holgado

Gonzalo Holgado

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Gonzalo Holgado has a total of 477 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Spectral classification of OB-type stars using tree-based ensemble methods and evaluation of their explainability

    by J. E. Gonzales, S. Simón-Díaz, S. Cuellar, J. A. Conejero, G. Holgado, A. de Burgos

    Context . The advent of large-scale spectroscopic surveys will deliver tens of thousands of spectra of blue massive stars spanning a wide range of spectral types (SpTs) and luminosity classes (LCs). This data volume renders traditional spectral classification techniques increasingly impractical and motivates the development of automated classification tools. Aims . We aim to develop a robust machine learning framework for the automated spectral classification of massive OB-type stars by assessing the performance and evaluating the interpretability of several tree-based ensemble algorithms. Methods . Using a large catalog of high-quality optical spectra of OB-type stars (plus a few A supergiants) compiled within the IACOB project, we designed three hierarchical experiments of increasing complexity: (1) classification into broad O, B, and A SpTs; (2) fine-grained classification within the O and B domains; and (3) joint classification of spectral subtype and LC. We evaluated six algorithms, ranging from single decision trees to ensemble methods (random forest and extra trees) and gradient boosting techniques (XGBoost, LightGBM, and histogram-based gradient boosting). We further analyzed model consensus and probabilistic outputs to assess the relative impact of epistemic and aleatoric uncertainties and employed Shapley additive explanations (SHAP) value analysis to provide an interpretation of the model decisions by identifying the features that drive each classification. Results . The models achieve excellent performance for broad SpT classification when compared with SIMBAD, with LightGBM reaching an accuracy above 98%. For fine-grained spectral subtypes, random forest provides the most robust results (89%). For the most challenging task combining SpT subtypes and LCs, XGBoost reaches a ∼ 77% accuracy. The decrease in performance is primarily driven by intrinsic degeneracies in the classical spectral classification scheme. Our SHAP analysis confirms that the models rely on meaningful spectral features. We also find a high level of agreement among different algorithms, indicating that current performance limits are largely set by the intrinsic complexity of the classification problem rather than by model choice. Conclusions . Tree-based ensemble methods provide a reliable and interpretable framework for the automated spectral classification of massive OB-type stars. The combination of full-spectrum modeling, algorithm comparison, and model interpretability offers a robust approach for future large-scale spectroscopic surveys.

  • Open Access

    SpecFANN: Spectral Fitting via Artificial Neural Networks

    by M. Abdul-Masih, C. Hawcroft, T. Lechien, S. Simon-Diaz, H. Sana, K. Deshmukh, J. Vrancken, J. I. Villaseñor, J. Müller-Horn, A. J. Kalita, et al.

    Context . The importance of massive stars cannot be overstated: they are powerful probes of the early universe, they play a vital role in the chemical and mechanical evolution of their host environments, and their end products allow us to study the most extreme physics in the universe. Obtaining accurate stellar and surface parameters for large samples of massive stars is vital to our understanding of how they evolve and how their births, lives, and deaths affect their surroundings. With the large volume of data expected from the next generation of spectroscopic surveys, it is likely that the computational cost of our current analysis methods (especially the calculation of model stellar atmospheres and synthetic spectra) will prove to be the most important bottleneck impeding our progress. Aims . In order to overcome these limitations and dramatically decrease computing times, we aim to develop a robust emulator for the FASTWIND radiative transfer and spectral synthesis code. Additionally, we aim to explore alternative fitting methods that have not been feasible up to this point due to computational costs. Methods . We calculated a large set (∼50 000 models in total) of FASTWIND synthetic spectra of OB-type stars, varying temperature, surface gravity, helium mass fractions, and CNOSi abundances, and we trained a collection of neural networks to emulate these models with separate networks for each line. We also developed the open-source python package SpecFANN (Spectral Fitting via Artificial Neural Networks), which provides users with a suite of fitting methods that can be used with these or other user-generated neural networks. Results . The majority of the trained neural networks reach average accuracies of better than ∼0.01–0.1% for photospheric lines and better than ∼0.1–1% for wind lines. We find that SpecFANN is able to obtain robust and accurate stellar parameters that are consistent with the literature for a sample of 52 early-type stars. Using SpecFANN we find that we can achieve the same fit in ∼1/360 000 of the time when compared to alternative techniques that rely on “on-the-fly” FASTWIND computations. Conclusions . We have demonstrated that neural networks offer a viable path forward to address the computational limitations of our current atmosphere analysis and stellar parameter determination methods for hot stars.

  • Open Access

    The IACOB project. XVIII. Prevalence of short-period binaries among Galactic He-rich O stars

    by C. Martínez-Sebastián, R. Gamen, O. G. Benvenuto, E. A. Saavedra, M. Carretero-Castrillo, M. Contreras, G. Holgado, J. Maíz Apellániz, S. Simón-Díaz

    For decades, the origin of helium enrichment in O-type stars has remained an open question. We investigate the correlation between surface helium abundance ( ) and orbital parameters for a sample of 45 O-type SB1 systems, including Cyg X-1. We find seven He-rich systems, all of which are concentrated at short orbital periods (P!łesssim!15 days) and are classified as RW. In addition, four of them present relatively high eccentricities, while the other three have ellipsoidal variations. We argue that these properties are the result of binary interaction. These findings provide strong observational evidence that binary interaction is the dominant origin of helium enrichment in O-type stars. This result has important implications for the treatment of chemical mixing and surface abundances in massive-star evolutionary models, and it establishes helium enrichment as a promising observational tracer for identifying post-interaction binary systems.

  • Open Access

    The IACOB project

    by C. Martínez-Sebastián, G. Holgado, S. Simón-Díaz, F. Martins, J. Puls

    Context. In recent years, a growing amount of evidence has revealed the crucial role of binarity in massive star evolution. This additional complexity compounds the uncertainties that still affect single-star evolution models and demands a refinement of the available observational constraints. A first crucial step toward this goal involves disentangling observed stars that have evolved in isolation from those that have experienced binary interaction. Aims. We aim to investigate the possible evolutionary origins of a sample of 117 Galactic O-type stars with luminosity classes V to III and projected rotational velocities ( v sin i ) below ∼150 km s −1 . Methods. We mostly focused on surface nitrogen and helium abundances but also considered other dynamical signatures that may help distinguish products of binary interaction from effectively single stars. We have therefore extended previous quantitative spectroscopic analyses performed within the framework of the IACOB project, and we obtained N abundance estimates. We investigated the correlations between these abundances and other stellar parameters, such as v sin i , effective temperature, surface gravity, and He abundance. As a reference, we used state-of-the-art predictions from single-star evolution models computed using different physical prescriptions. Results. We found good agreement between our N abundance estimates and previous determinations based on different analysis methodologies and stellar atmosphere codes. We identified clear differences in the N abundance distributions corresponding to three He abundance regimes, defined as He-low ( Y He = N (He)/ N (H)≤0.08), He-normal (0.08 < Y He ≤ 0.12), and He-rich ( Y He > 0.12). We argue that the abundance estimates for the He-low group, as well as for some additional stars with abnormally low N abundances, are likely spurious determinations. For the He-normal group, the N abundance distribution peaks slightly above the expected birth value and extends up to ϵ N = log(N/H)+12 ∼ 8.4 dex. For these stars, we found an overall agreement with single-star evolutionary models that include efficient internal mixing and assume moderate-to-low initial rotation ( v ini / v crit ≲ 0.2). In contrast, the He-rich group exhibits a bimodal N abundance distribution, with one peak at ∼8.1 dex corresponding to mildly enriched stars and a second more enriched peak around ∼8.5 dex. None of these stars are consistent with predictions from state-of-the-art single-star evolutionary models. Conclusions. We argue that the two N abundance subgroups among the He-rich stars are most plausibly explained as binary products. Furthermore, despite the N abundance in He-normal stars with luminosity classes (LC) IV and V being reproduced by single-star evolutionary models with efficient mixing models, the same models predict a higher N abundance than observed for stars in this group with LC III. This indicates that rotational mixing alone is unable to explain the observed distribution of N abundances among stars with normal He abundances. A future comprehensive study of surface abundances in O-type supergiants (LC I and II) and fast rotators that also incorporates the abundances of additional elements is essential to further constraining the evolutionary channels of the most massive stars during the main sequence.


  • Open Access

    The IACOB project

    by S. Simón-Díaz, G. Holgado, C. Martínez-Sebastián, M. Carretero-Castrillo, H. Jin, M. A. Urbaneja, R. Gamen, J. Puls, A. de Burgos, M. Garcia, et al.

    The presence of massive O-type stars with surfaces enriched by CNO-cycle products has been known since the early 1980s. For many years, internal rotational mixing was assumed to be the dominant mechanism responsible for this chemical contamination. However, accumulating evidence suggests that binary interaction may play an equally important, if not dominant, role.


  • Open Access

    Binarity at LOw Metallicity (BLOeM)

    by D. J. Lennon, S. R. Berlanas, A. Herrero, N. Britavskiy, P. L. Dufton, N. Langer, H. Jin, A. Schootemeijer, A. Menon, J. Bestenlehner, et al.

  • Open Access

    The OWN Survey. A high-resolution spectroscopic survey of southern Galactic O and WN-type stars

    by R. H. Barbá, R. Gamen, N. I. Morrell, J. I. Arias, J. Maíz Apellániz, R. Higa, C. Rodriguez, C. Putkuri, G. A. Ferrero, T. Ansín, et al.

    Context. Massive stars play crucial roles in galactic dynamics and chemical evolution. They are the most significant sources of ionizing UV radiation, their substantial mass-loss rates and explosions inject energy and enrich their surroundings, and their dynamical interactions eject stars and alter the evolution of stellar clusters. Consequently, the study of massive stars is essential for understanding various astrophysical phenomena, including galaxy chemical evolution, interstellar medium dynamics, gamma-ray bursts, and the reionization of the Universe. Key parameters influencing the evolution of massive stars include mass, mass-loss rate, chemical composition, and rotation. The orbits of spectroscopic binaries are particularly valuable because they provide constraints on stellar masses, and when combined with complementary data (e.g., photometry or interferometry), these masses can be fully determined. Aims. The OWN Survey was started two decades ago to study Galactic O- and WN- (hence the name) type southern spectroscopic binaries. In this paper we present the final results for single-lined (SB1) spectroscopic orbits. Methods. The OWN Survey carried out a long-term spectroscopic campaign to search for radial velocity variations indicative of orbital motion in a sample of southern Galactic O- and WN-type stars with high-resolution spectrographs in Argentina and Chile. The OWN spectra were later combined with high-resolution spectra from other sources and, in some cases, photometric time series to derive orbits and disentangled spectra, from which masses were constrained or determined and spectral classifications obtained. High-resolution optical spectra of 212 massive stars were obtained during the ∼20 years of the OWN project, and each target was observed at least three times. Results. Among the 212 stars, 144 exhibited radial-velocity variations greater than 15 km s −1 . We present a complete and coherent compilation for the 23 systems with single-lined spectroscopic orbits identified in our sample. In Paper II we will perform a similar analysis for the systems with double-lined spectroscopic orbits.


  • Open Access

    Binarity at LOw Metallicity (BLOeM)

    by D. J. Lennon, S. R. Berlanas, A. Herrero, N. Britavskiy, P. L. Dufton, N. Langer, H. Jin, A. Schootemeijer, A. Menon, J. Bestenlehner, et al.

    The Binarity at LOw Metallicity (BLOeM) survey is an ESO large programme designed to obtain multi-epoch spectroscopy for 929 massive stars in the Small Magellanic Cloud (SMC). It will provide binary fractions and orbital configurations of binary systems and search for dormant black hole binary candidates (OB+BH). We present projected rotational velocities ( v sin i ) of all sources, and, using the multiplicity properties presented in previous papers, we derive the v sin i distributions of apparent single stars, single-lined spectroscopic (SB1) binaries, and SB2 systems. We identify a locus in the Hertzsprung-Russell diagram where rotational velocities decrease significantly; we interpret this feature as broadly corresponding to the terminal-age main sequence. The main-sequence cohort is distinguished by a broad range of v sin i values, but with a strong peak in the distribution in the range 30–60 km s −1 , which is close to the resolution limit of ∼30 km s −1 , indicating the presence of many upper limits. Sources in this low v sin i peak are distributed throughout the main sequence and are also present in the SB1 sample, though less prominent than in the single-star distribution. A preliminary analysis of the lowest v sin i cohort, which includes SB1 systems, implies that roughly one-third may be nitrogen rich, and we speculate that this cohort is a mix of pristine single stars, long-period binaries, and merger products. The SB2 systems appear to be mostly short-period binaries in synchronous rotation, and their v sin i estimates are distributed around a mean value of ∼140 km s −1 . Higher v sin i sources are also present in the single and SB1 systems, all of which have a tail to higher v sin i values. This is consistent with tidal and mass-transfer effects. The supergiants, with a few exceptions, have low v sin i , and the bulk of these systems is essentially unresolved at the current spectral resolution (∼30 km s −1 ).


  • Open Access

    An observational study of rotation and binarity of Galactic O-type runaway stars

    by M. Carretero-Castrillo, M. Ribó, J. M. Paredes, G. Holgado, C. Martínez-Sebastián, S. Simón-Díaz

    Context.Gaia Data Release 3 (DR3) has revealed new massive runaway stars and large spectroscopic surveys have now enable detailed characterization studies. However, the relative contributions of binary supernova (BSS) and dynamical ejection (DES) scenarios to explain their runaway origin remain poorly constrained, particularly in the Milky Way. Aims. We aim to characterize the largest sample of Galactic O-type runaway stars ever investigated through their kinematics, rotation, and binarity with the ultimate objective of shedding light on their potential runaway origins. Methods. We used the GOSC- Gaia DR3 catalog of normal and runaway stars, along with IACOB spectroscopic information to build a sample with 214 O-type stars with information on the projected rotational velocities ( v sin i ). We also built a subsample of 168 O-type stars with additional information about their likely single (LS) or single-lined (SB1) spectroscopic binary nature. We also considered an additional sample of 65 double-lined (SB2) spectroscopic binaries. Results. We find that among our sample of Galactic O-type runaways, most (74%) have v sin i < 200 km s −1 , whereas for normal stars this fraction is slightly higher (82%). There are no fast-moving ($ V_{\mathrm{PEC}}^{\mathrm{2D}} > 85 $ km s −1 ) runaways that have been shown to be fast rotators ( v sin i ≥ 200 km s −1 ), except for HD 124 979 . Runaways exhibit lower SB1 fractions than normal stars, with no runaway SB1 fast-rotating systems; on average, runaways rotate faster than normal stars; and their runaway fraction is higher among fast rotators (44%) versus the slow rotators (34%). This is consistent with BSS dominance for fast rotators. We also found that SB2 systems hardly reach runaway velocities with a low runaway fraction (10%). Runaways with $ V_{\mathrm{PEC}}^{\mathrm{2D}} > 60 $ km s −1 are mostly single and interpreted as DES products, while runaways with $ V_{\mathrm{PEC}}^{\mathrm{2D}} > 85 $ km s −1 are also interpreted as two-step products, with the binary V479 Sct / LS 5039 a likely example. Finally, we found that three of 12 runaway SB1 systems are high-mass X-ray binaries. Conclusions. Our observational study reveals that Galactic O-type runaways are dominated by slow rotators. The study points to a dominance of BSS among fast-rotating runaways and of DES and the two-step scenario among the high-velocity ones. The observed trends provide valuable constraints on models on the origins of runaway stars.


  • Open Access

    Predicting Stellar Parameters of Massive Stars from Light Curves with Machine Learning

    by Rachel C. Zhang, Kaze W. K. Wong, Gonzalo Holgado, Matteo Cantiello

    High-resolution spectroscopic measurements of OB stars are important for understanding processes like stellar evolution but require labor-intensive observations. In contrast, photometric missions like the Transiting Exoplanet Survey Satellite (TESS) can monitor hundreds of thousands of stars with a range of temporal resolutions but do not provide such detailed measurements. With surveys like the Legacy Survey of Space and Time promising unprecedented photometric coverage over the next 10 yr, it is increasingly important to develop methods that connect large-scale time-series photometry with the detailed stellar parameter measurements typically derived from spectroscopy. In this paper, we test whether machine learning can recover such parameters by combining TESS light curves with spectroscopic measurements from the IACOB project, using a sample of 285 light curves from 106 unique O stars. Using both multilayer perceptrons and convolutional neural networks, we demonstrate that (1) O star light curves contain sufficient information to meaningfully infer stellar parameters, and (2) periodograms derived from light curves capture substantially more information than previously identified correlation parameters. Our best model achieves moderate success in predicting both spectroscopic luminosity ( R 2 = 0.64 1 − 0.167 + 0.107 ) and effective temperature ( R 2 = 0.44 3 − 0.234 + 0.056 ), key stellar parameters for determining positions of stars on the spectroscopic Hertzsprung–Russell diagram, despite the small dataset size. Further progress will require expanded datasets of matched photometric and spectroscopic observations.