First and Comprehensive Study of V0757 Pup: gamma- Doradus Pulsator in Detached Eclipsing Binary

arXiv:2608.11896 · astro-ph.SR · Submitted 2026-08-12 · Read on arXiv

The Graduate University for Advanced Studies, SOKENDAI · National Astronomical Observatory of Japan · Shanghai Jiao Tong University · Institut Teknologi Bandung · Bosscha Observatory · Institut Teknologi Sumatera · National Astronomical Research Institute of Thailand

astro-ph.SR

Submitted: 2026-08-12

Updated: 2026-08-12

Comments: 22 pages, 15 figures, submitted to PASJ. Comments are welcome

Code: https://github.com/ich-heisse-eugene/PyYAP

Project page: https://lightkurve.github.io/lightkurve/index.htmlhttps://trt.narit.or.th/3https://www.astro.louisville.edu/software/astroimagej/2

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

Importance score: 75/100

The gist: This paper presents the first comprehensive study of V0757 Pup (TIC 6939791), a detached eclipsing binary system containing a γ Doradus pulsator.

Terminology

Summary

This paper presents the first comprehensive study of V0757 Pup (TIC 6939791), a detached eclipsing binary system containing a γ Doradus pulsator. The system consists of an F2V primary star (M1 = 1.305 ± 0.026 M⊙, R1 = 1.643 ± 0.020 R⊙) and a G1V secondary star (M2 = 0.934 ± 0.030 M⊙, R2 = 0.941 ± 0.079 R⊙). The analysis combines high-precision photometry from four TESS sectors (Sectors 7, 34, 61, and 88) with ground-based medium-resolution spectroscopy from the Thai National Telescope (MRES spectrograph) and photometric follow-up from the Thai Robotic Telescopes network.

The orbital ephemeris was determined from 100 eclipse minima (51 primary and 49 secondary) spanning over 24 years, yielding an orbital period of Porb = 1.9891150352 ± 0.0000003845 days. Eclipse Timing Variation analysis revealed no significant deviations due to a third body, and Gaia astrometry showed a proper motion anomaly significance of only 2.56σ, below the 3σ detection threshold.

Spectroscopic analysis using Least Squares Deconvolution and spectral disentangling determined the primary's effective temperature Teff = 7033+54−80 K, surface gravity log g = 3.988+0.105−0.104, and projected rotational velocity v sin i = 41.201 ± 0.031 km/s. The interstellar extinction was measured using Na I D doublet absorption lines, yielding E(B − V) = 0.025+0.005−0.004 mag, consistent with other estimates.

Simultaneous radial velocity and light curve modeling using the Wilson-Devinney code (PyWD2015) provided precise absolute parameters. The derived photometric distances (387.94+54.21−45.01 pc in V band, 357.38+50.61−41.58 pc in G band) agree with the Gaia DR3 parallax distance of 352.373 ± 2.136 pc, and SED fitting gave dSED = 349.59+2.13−2.08 pc.

Frequency analysis of the residual light curves identified two dominant independent pulsation frequencies at f ∼ 0.79 c/d and 0.98 c/d, persistent across multiple TESS sectors. The derived pulsation constants (Q ≃ 0.5–0.6 days) confirm the primary component as a γ Doradus pulsator oscillating in g-modes. No solar-type oscillations were detected in the secondary component.

Evolutionary modeling using MIST isochrones indicates the system is approximately 2.2 Gyr old. The primary component is expected to fill its Roche lobe in about 800 Myr (at a stellar age of 3.04 Gyr), at which point the system will transition from a detached to a semi-detached binary. A systematic metallicity offset was noted between spectroscopic analysis ([Fe/H]spec ≈ −0.625 dex) and isochrone fitting (higher by 0.2 dex), potentially explained by 3D and non-LTE effects on Fe lines.

Improvements for AI systems

Improvements to AI Systems:

  1. Automated Pulsation-Component Classification in Eclipsing Binaries
  • Train a model on multi-sector TESS light curves to automatically separate eclipse signals from residual pulsation frequencies (e.g., using Gaussian process regression or iterative pre-whitening). The improved AI can classify a star as a γ Doradus pulsator within a binary without manual frequency extraction, outputting pulsation constants (Q) and mode identification (g-modes) in real time.
  1. Unified Spectroscopic-Photometric Parameter Inference
  • Develop a neural network that jointly fits radial velocities, light curves, and spectral disentangling outputs (from LSD) to derive stellar masses, radii, and temperatures simultaneously, reducing systematic biases. The improved AI can produce a full set of absolute parameters (M, R, Teff, log g, v sin i) for any detached binary with a pulsating component, with uncertainty propagation from all data sources.
  1. Eclipse Timing Variation (ETV) Anomaly Detector with Astrometric Cross-Validation
  • Implement a transformer-based model that ingests eclipse minima times and Gaia proper motion anomalies to predict the presence of a third body. The improved AI can flag sub-3σ signals (like the 2.56σ here) as non-significant, avoiding false positives, and automatically compute the detection threshold for any binary system.
  1. Metallicity Offset Corrector for Isochrone Fitting
  • Create a correction module that learns the systematic offset between spectroscopic [Fe/H] and isochrone-derived metallicity (here 0.2 dex) as a function of Teff, log g, and line depth, using 3D non-LTE synthetic spectra. The improved AI can output a calibrated metallicity for any F/G star, improving age and evolutionary stage estimates (e.g., 2.2 Gyr) by up to 20% accuracy.
  1. Automated Roche-Lobe Overflow Time Predictor
  • Train a regression model on MIST isochrone grids and binary parameters (masses, radii, period) to predict the time until a primary fills its Roche lobe. The improved AI can instantly forecast the transition from detached to semi-detached status (e.g., 800 Myr here) for any binary, enabling large-scale surveys of evolving systems.
  1. Multi-Sector Frequency Persistence Classifier
  • Build a deep learning model that analyzes residual light curves across multiple TESS sectors to identify stable pulsation frequencies (like f = 0.79 and 0.98 c/d) and reject transient noise. The improved AI can output a persistence score for each frequency, automating the selection of genuine g-modes in long-baseline photometry.
  1. Distance Estimator Fusing Photometric, SED, and Gaia Parallax
  • Implement a Bayesian neural network that combines V-band, G-band, SED, and Gaia DR3 parallax distances (as in this paper) to produce a single, most probable distance with correlated uncertainties. The improved AI can reduce distance errors by 15% for field stars with extinction measurements (E(B−V) from Na I D lines) and flag inconsistencies.

What the improved AI system can do:

  • Given raw TESS light curves, spectra, and Gaia data for any detached eclipsing binary, it can automatically output: (1) orbital and stellar parameters, (2) pulsation classification and frequencies, (3) third-body detection probability, (4) calibrated metallicity and age, (5) predicted Roche-lobe filling time, and (6) a fused distance—all without manual intervention, with uncertainties comparable to this study’s precision. It can also scale to thousands of systems in surveys like TESS or PLATO, enabling statistical studies of binary evolution and pulsator demographics.

Abstract

Pulsating stars in detached eclipsing binary (EA) systems are known for providing many important physical informations which constrain both stellar structure and evolution theories. To date, fewer than 50 gamma Doradus (GDOR) stars have been found in eclipsing binary systems, making them important targets to study g-mode pulsation inside the stars. We present a comprehensive physical and pulsational analysis of V0757 Pup (TIC 6939791), a detached eclipsing binary system observed by the TESS mission (Sectors 7, 34, 61, and 88) and followed up with ground-based spectroscopy from the Thai National Telescope. By combining light curve modeling with radial velocities derived from medium-resolution spectra, we determined the fundamental stellar and atmospheric parameters with high precision. The system consists of an F2V primary (M 1=1.305 plus or minus0.026M, R 1=1.643 plus or minus0.020R) and a G1V secondary (M 2=0.934 plus or minus0.030M, R 2=0.941 plus or minus0.079R). We performed a detailed frequency analysis of the residual light curves, identifying two dominant independent pulsation frequencies at f about0.79 c/d and 0.98 c/d. These frequencies, along with the derived pulsation constant (Q about0.6 d), confirm the primary component as a gamma Doradus pulsator. An analysis of Eclipse Timing Variations (ETV) reveals non detection of third body companions with current available dataset, strengthen by the result of Gaia astrometric analysis. Evolutionary modeling indicates the system is about2.2 Gyr old, with the primary expected to fill its Roche lobe in about1 Gyr. Additionally, we calculate a distance of d about350 pc based on orbital and SED modeling, which is in excellent agreement with the Gaia DR3 parallax distance.

Sources

Related papers