The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry

arXiv:2608.03383 · astro-ph.SR · Submitted 2026-08-04 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: Next we'll be talking about the paper "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry".

Jocelyn: The paper was written by K. Thomson-Paressant et al. from.

Vera: Stay tuned as we take you through the paper and discuss its implications.

Summary: Vera: We’ve seen how big the sample is, but what does "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry" actually tell us about these diverse stellar behaviors?

Jocelyn: The summary reveals that variability is incredibly common; they found over ninety-three percent of the sample exhibits some form of photometric variation.

Subrahmanyan: That high prevalence confirms our intuition that stellar properties in this mass range are rarely simple, and it suggests we need to prepare our theoretical models for a highly complex environment.

Vera: And when looking specifically at pulsations, they found that eighty-two percent of the sample shows signs of these oscillations, mostly from types like beta Cep and slowly pulsating B-type stars.

Jocelyn: It's interesting that the observations show such a high percentage of pulsators; it’s almost as if we are seeing a natural frequency in how these stars behave.

Subrahmanyan: This high fraction of pulsators is significant because those pulsations are direct probes into the core and envelope, letting us map out internal structures that would otherwise be inaccessible.

Vera: The data also showed that at least fourteen percent of the stars have evidence of binarity, which is a major finding when combining both spectroscopic and photometric clues.

Jocelyn: That number is quite high, suggesting that in many cases where we think we see a single star, there might be an entire system working beneath our view.

Subrahmanyan: Multiplicity is a fundamental characteristic of massive stars, and this data confirms that these systems are often the dominant actors in galactic dynamics.

Improvements: Vera: We’ve established that variability is widespread, but "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry" offers some really important methodological improvements for how we analyze these stars.

Jocelyn: The authors used very detailed cross-correlation techniques on the normalized spectra to measure radial velocity variability, which is a major step up from simple visual inspection.

Subrahmanyan: That refined measurement of RV variability is crucial because it allows us to separate physical motions—like orbital wobble—from other forces that influence the star’s center of mass.

Vera: And they didn're not just using any old standard; they employed a sophisticated machine-learning technique, SUPPNet, for continuum normalization across the HERMES data.

Jocelyn: That automation is a huge win for efficiency; it makes it practical to apply these high standards across such an extensive sample of eight hundred seventy-three stars.

Subrahmanyin: This robust parameter estimation helps us place these stars accurately on the Hertzsprung–Russell diagram, which is where we can really test our evolutionary models against the observed data.

Vera: The paper has provided a statistical framework for future studies, suggesting that this work isn't just a collection of observations, but a blueprint for future research.

Jocelyn: It sets a standard for how we should be approaching massive star surveys in the Northern Hemisphere, complementing those efforts in other parts of the next survey.

Methodology Deep Dive: Vera: Since we’ve talked about the results, let’s talk about how they did it—the methodology within "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry."

Jocelyn: They used two distinct sets of data—TESS light curves and multi-epoch HERMES spectra—and then they cross-referenced the results to find systems that might be both pulsating and binary.

Subrahmanyan: The way they use TLUSTY model grids to fit the observed spectra gives us a very precise estimate of things like effective temperature and surface gravity, which is much more robust than just using observational estimates.

Vera: They also had to deal with the fact that not every star has three epochs, so they used specific criteria to identify a candidate binary system based on RV variability.

Jocelyn: That's where they defined two very strict criteria—first, a statistically significant jump in radial velocity, and then RV must be larger than a threshold of twenty km s−one.

Subrahmanyan: But this brings up the challenge because this threshold is so high; it essentially filters out most non-binary RV variability, which is where the overlap with pulsations comes into play.

Vera: The data showed that only about one per cent of stars were both non-pulsating and showing significant RV variability, which is a really important distinction.

Jocelyn: It seems like the authors are being very cautious and conservative when classifying these systems, ensuring we don't mislabel pulsators as binaries.

Conclusion & Wrap-up: Vera: So, looking at "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry" as a whole, what is the ultimate impact on our understanding these stars?

Jocelyn: It’s a giant leap forward in inventory and classification; we have a massive database that gives us empirical evidence for the prevalence of both pulsations and binarity across the entire population.

Subrahmanyan: The ability to use this large, homogeneous sample allows us to test theoretical models of stellar structure against observed variability with a level of detail we never had before.

Vera: I think the biggest message is that we need to be more nuanced; given the limited number of epochs for most stars, it’s clear that using a single threshold for binarity might miss many real systems.

Jocelyn: That caution is key; we must continue monitoring these targets with more follow-up spectroscopy to fully understand the relationship between one system and its true binary nature.

Subrahmanyan: The results suggest that the mechanisms governing stellar evolution—like internal mixing—are highly sensitive to mass, which will be tested against these observed trends.

Vera: We're really looking forward to seeing how this data is used in future studies, especially when the next generation of missions like PLATO comes online.

Jocelyn: I agree; it provides a strong foundation for the next step in the survey work.

Subrahmanyin: It gives us a powerful way to connect observed stellar behavior back to the big picture cosmic processes we are trying to understand.

K. Thomson-Paressant et al.

astro-ph.SR

Submitted: 2026-08-04

Updated: 2026-08-04

Comments: 24 pages, including 7 of appendices, 15 figures, 2 tables

Code: https://github.com/TomerShenar/Cross-correlation

Project page: https://archive.stsci.edu/missions-and-data

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 86/100

The gist: The scientific paper "The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry" presents a detailed study of stellar variability in a large population

Key concepts

Photometric Variation: This refers to the measurable changes in brightness observed in TESS data. The study found that over 93% of the massive star sample exhibits some form of this variation, indicating that stellar properties in this mass range are rarely simple.
Pulsations: These are oscillations within a star's structure, such as those seen in B-type stars. The study found 82% of the sample shows signs of these oscillations, which is significant because they allow researchers to map out internal structures inaccessible through other direct probes.
Binarity: This refers to a star being part of a binary system where two objects orbit each other. At least 14% of the stars in the study showed evidence of binarity, suggesting that many systems thought to be single might be an entire system working beneath our view.

Terminology

Metallicity — 8× in this paper · explained in 3 episodes
Core — 4× in this paper · explained in 2 episodes
Consistency — 2× in this paper · explained in 3 episodes
European Research Council — 2× in this paper
Accuracy — 1× in this paper · explained in 1 episode
Actor — 1× in this paper · explained in 3 episodes
Alignment — 1× in this paper · explained in 1 episode
Aperture — 1× in this paper · explained in 1 episode
Calibration — 1× in this paper · explained in 7 episodes
Mikulski Archive for Space Telescopes — 1× in this paper

Summary

The scientific paper The symphony of pulsations and binarity among massive stars using HERMES spectroscopy and TESS photometry presents a detailed study of stellar variability in a large population of O- and B-type stars.

Objective and Methodology

The study aims to investigate the prevalence and interplay between different types of variability—including binarity, pulsation, and rotation—among massive stars. The researchers analyze data from a large population of 873 O- and B-type stars located in the Northern hemisphere. The methodology involves a combined photometric and spectroscopic analysis:

Photometric Analysis: Time series photometry from TESS are analyzed using standard frequency analysis methods to detect pulsations and rotational modulation signatures.
Spectroscopic Analysis: High-resolution HERMES spectra are used to identify radial velocity (RV) variability. The spectroscopic data is normalized using machine-learning techniques (SUPPNet) and compared against a grid of synthetic TLUSTY spectra to determine stellar parameters, such as effective temperature (T eff), surface gravity (g), and projected rotational velocity (v i).

Key Findings on Variability

The analysis reveals that variability is extremely common in this population:

Photometric Variability: More than 93 per cent of the sample exhibits photometric variability.
This variability is caused by pulsations in 82 per cent of the sample. The dominant contributions are identified as beta Cep and slowly pulsating B-type stars, along with stochastic low-frequency (SLF) variability.
Binarity: Based on limited spectroscopic epochs, at least 14 per cent of the stars show evidence of binarity, encompassing both eclipsing and spectroscopic systems.

Detailed Ensemble Results

The study provides a comprehensive breakdown of the classifications for the 873 stars:

Pulsators: A total of 664 pulsating stars were identified. This includes:
197 stars with g-mode pulsations characteristic of SPB (slowly pulsating B-type) stars.
251 stars with p- and/or g modes typical of beta Cep stars.
268 stars showing SLF variability. This represents a pulsator fraction exceeding 80 per cent, which is consistent with previous studies.
Eclipsings Binaries (EBs): The TESS light curves identified 67 eclipsing binaries (EBs), including 30 new EBs not previously known in the literature.
Rotational Modulation: 148 stars were identified as showing rotational modulation, suggesting potential large-scale magnetic fields.
Spectroscopic Binaries: Out of the 873 stars, 353 show RV variability (or 76 per cent of the 464 stars with at least two spectra). Of these, 51 satisfy the two statistical criteria for binarity.

Discussion on Stellar Parameters and Models

The study compares parameters derived from TLUSTY models with those obtained from the Gaia DR3 archive:

The researchers found that while T eff values are largely consistent between methods, discrepancies in g and v i are common.
The comparison demonstrates that the TLUSTY grid-based fitting method is reliable, showing consistency with manual normalization via the use of the SUPPNet automated technique.

Conclusion and Significance

This work constitutes one of the largest homogeneous surveys of variability for intermediate-mass and massive stars in the Northern hemisphere, complementing similar efforts in the Southern hemisphere. It provides a statistical framework for future studies of stellar structure and evolution, particularly in the context of asteroseismology.

Improvements for AI systems

I must approach this with extreme caution; the complexity of stellar astrophysics requires highly specialized AI components to avoid catastrophic misinterpretations of physical parameters.

Based on the methodologies presented—specifically the comparison of continuum normalization techniques (manual vs. SUPPNet) and the detailed analysis of photometric light curves from eclipsing binaries—three major advancements are necessary for robust, high-throughput AI systems.

Improvement: Develop a specialized deep convolutional neural network (CNN) architecture, potentially incorporating Variational Autoencoders (VAEs), trained explicitly on synthetic spectra generated by stellar atmosphere codes (like TLUSTY). This system must move beyond simple curve fitting to learn the underlying physical continuum structure and simultaneously predict stellar parameters (T eff, g).

Implementation Detail: The CNN should be trained not just on the observed flux, but on the residuals after a theoretically derived continuum level is subtracted. A secondary module should implement an attention mechanism to weigh spectral regions most sensitive to specific atmospheric features (e.g., Balmer lines for g, metal lines for T eff).

What the Improved AI System Can Do:

Automated High-Fidelity Normalization: It will perform continuum normalization with reliability exceeding manual methods, drastically reducing the systematic uncertainties associated with normalization residuals (delta F/F).
High-Throughput Stellar Characterization: It can analyze entire stellar surveys (hundreds of thousands of spectra) to extract T eff and g for massive stars in minutes, flagging outliers or regions where the spectral quality is insufficient, thereby minimizing human intervention.
Systematic Bias Detection: By training on known systematic errors (e.g., detector response variations, telluric contamination), the system can proactively quantify and correct these biases across an entire dataset.

Abstract

A wide range of variability mechanisms exist among intermediate mass and massive stars, which are not yet fully understood. Using complementary data sources for a large population of B- and O-type stars, we aim to study the prevalence and interplay of different types of variability, including binarity, pulsation, and rotation, to prepare for future modelling. To this end, we analyse high-resolution HERMES spectra and 2-min cadence TESS photometry and characterise the diverse variability observed within a population of 873 O- and B-type stars. The spectroscopic data were normalised using machine-learning techniques, compared to a grid of synthetic TLUSTY spectra to determine stellar parameters, and used to identify radial velocity variability. Photometric time series were analysed using standard frequency analysis methods to detect pulsations and rotational modulation signatures. We find that more than 93 per cent of the sample exhibits photometric variability. Photometric variability caused by pulsations is identified in 82 per cent of the sample, with dominant contributions from beta Cep and slowly pulsating B-type stars, as well as stochastic low-frequency variability. Based on a limited number of spectroscopic epochs, at least 14 per cent of the stars show evidence of binarity, including both eclipsing and spectroscopic systems. This work represents one of the largest homogeneous surveys of variability for intermediate-mass and massive stars in the Northern hemisphere, and complementing similar efforts in the Southern hemisphere. It provides a statistical framework for future studies of stellar structure and evolution, particularly in the context of asteroseismology.