Astrometric modeling of unresolved variable binary systems

arXiv:2611.14690 · astro-ph.SR, astro-ph.IM · Submitted 2026-11-10 · Read on arXiv

Listen

Radio episode about this paper

Transcript

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

Vera: We're looking at the paper: "Astrometric modeling of unresolved variable binary systems".

Jocelyn: Here's the gist of the paper. A framework is developed to simulate and interpret epoch astrometry of unresolved binary systems and evaluate the impact of model assumptions on resulting astrometric parameters The new vimbinary tool is introduced to simulate Gaia DR4-like epoch astrometry, with particular emphasis on VIM effects Context and Motivation The fourth data release of the Gaia mission (DR4) will provide epoch astrometry for the first time, necessitating careful treatment of binary sources as modeling them as single stars can bias parallax and proper motion estimates.

Vera: The paper aims to develop a framework to simulate and interpret epoch astrometry of unresolved binary systems and evaluate the impact of model assumptions on the resulting astrometric parameters for the systems considered here.

Jocelyn: Modeling Gaia Epoch Astrometry The simulation involves modeling along-scan (AL) measurements, which are time-dependent quantities consisting of a time series of along-scan positions with respect to the reference position adopted in the global astrometric solution.

Vera: The single-star astrometric signal is expressed as wss = (∆α⋆ + µα⋆ t) sin θ + (∆δ + µδt) cos θ + Πϖϖ.

Paper summary: Vera: To go deeper into what the paper actually claims, this work by Sivkova and Kervella is setting up a framework to model along-scan measurements as time-dependent quantities. They start by expressing the single star astrometric signal with terms involving the change in position, proper motion, and parallax.

Jocelyn: And then they build on that by modeling binary signals using a hierarchy of complexity. They use fixed models like VIMF, then move to linear motion like VIML and accelerated motion like VIMA, all the way up to orbital models called VIMO for variability induced movers or VIMs.

Subrahmanyan: The core concept they are testing is how these different assumptions about the nature of the system—single star versus various binary models—translate into different astrometric parameters, particularly when dealing with the variability component.

Vera: They spend a lot of time discussing how both binary motion and the VIM effect can significantly impact parallax and proper motion measurements. They emphasize that the magnitude of this impact really depends on the orbital period, the photocentric amplitude, and the flux ratio specific to each system they look at.

Jocelyn: It’s interesting because they note a key difference: larger time coverage doesn't necessarily improve parallax precision, but it does help with proper motion precision for those longer baselines. They also highlight that parallaxes retrieved using binary-system fitting are generally more accurate than those obtained within the single-star framework.

Subrahmanyan: For the bigger picture, this means that our way of understanding stellar distances from Gaia needs to incorporate these modeling assumptions correctly, because if we don't account for the companion's effect on the light curve, our distance calibrations are fundamentally flawed.

Vera: They validate their entire framework by comparing their simulations against existing datasets like Gaia BH3, and they achieved a chi squared red of zero point six five when checking the simulated along-scan positions against those published measurements in terms of noise level.

Jocelyn: And one specific result they highlighted was that proper motion correction for unresolved binaries is necessary because subtracting the fitted proper motion from the Gaia DR3 proper motion gives an estimate closer to the true barycentric proper motion.

Subrahmanyan: This confirms that there's a systematic offset in how we calculate those motions when you try to disentangle a binary system, which is important context for any follow-up observations we might plan.

Conclusion: Vera: So wrapping up "Astrometric modeling of unresolved variable binary systems," the authors are presenting this vimbinary tool as a Python package designed to model Gaia DR4 like epoch astrometry, usable both before and after the release.

Jocelyn: The real implication is that accounting for these VIM effects with an appropriate model is expected to improve the calibration of the period-luminosity relation with Gaia DR4, both directly and through a better characterization of the projection factor used in the parallax-of-pulsation method.

Subrahmanyan: They also suggest that variability-induced motions are likely to become more important for asymptotic giant branch stars because those stars have large photometric variability, meaning convection induced variations will strongly affect the astrometric solution for them.

Vera: And they conclude that for systems with small photocentric orbits, binary motion doesn't introduce a significant bias in the parallax, but they stress that the scatter in parallax for Cepheid systems remains significant across different releases.

Jocelyn: So the final message is that for binary Cepheids, individual case analysis is more suitable to achieve the best parallax precision and you need to look closely at how these effects play out depending on what you're looking at.

LIRA Observatoire de Paris University PSL Sorbonne University University Paris Cité CY Cergy Paris University CNRS · French-Chilean Laboratory for Astronomy IRL CNRS U de Chile · Instituto de Alta Investigación Universidad de Tarapacá Casilla 7D Arica Chile · Institute of Astronomy KU Leuven Celestijnenlaan 200D Leuven Belgium · Université Côte d’Azur Observatoire de la Côte d’Azur CNRS Laboratoire Lagrange Nice France · UNIDIA Observatoire de Paris University PSL CNRS Meudon France · European Southern Observatory Headquarters Karl-Schwarzschild-Str. 2 Garching Germany · Smithsonian Astrophysical Observatory MS 4 Garden Street Cambridge MA USA · European Space Agency ESA Office Space Telescope Science Institute Baltimore MD USA

astro-ph.SR, astro-ph.IM

Submitted: 2026-11-10

Updated: 2026-11-10

Comments: Accepted for publication in A&A. Code available at https://github.com/katsivkova/vimbinary

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 91/100

The gist: A framework is developed to simulate and interpret epoch astrometry of unresolved binary systems and evaluate the impact of model assumptions on resulting astrometric parameters The new vimbinary…

Key concepts

Epoch Astrometry
This refers to measuring the precise positions of a star over long periods using data from different missions, like Gaia. For binary systems, this involves tracking how their combined position changes over time while accounting for the orbital motion and variability.
VIM Effects
Variability-induced motions are movements in a binary system caused by changes in brightness or flux. These effects can significantly bias astrometric measurements like parallax and proper motion, requiring specific models to accurately interpret the observed data from missions like Gaia.
Along-Scan (AL) Measurements
These are time-dependent measurements of a star's position along a specific line of sight in the sky. They are used to model how the system's position changes over time relative to a reference point in the global astrometric solution.
Vimbinary Tool
This is a Python package designed to simulate Gaia DR4-like epoch astrometry for binary systems. It allows researchers to test different models, such as fixed, linear, or orbital models, and evaluate how assumptions about the binary's motion affect the final calculated astrometric parameters.

Terminology

Summary

A framework is developed to simulate and interpret epoch astrometry of unresolved binary systems and evaluate the impact of model assumptions on resulting astrometric parameters The new vimbinary tool is introduced to simulate Gaia DR4-like epoch astrometry, with particular emphasis on VIM effects

Context and Motivation

The fourth data release of the Gaia mission (DR4) will provide epoch astrometry for the first time, necessitating careful treatment of binary sources as modeling them as single stars can bias parallax and proper motion estimates. The paper aims to develop a framework to simulate and interpret epoch astrometry of unresolved binary systems and evaluate the impact of model assumptions on the resulting astrometric parameters for the systems considered here. p. 0

Where the paper says it

When analysing these data, binary sources must be treated carefully, as modelling their astrometry as single stars can bias parallax and proper motion estimates. We aim to develop a framework to simulate and interpret epoch astrometry of unresolved binary systems and to evaluate the impact of model assumptions on the resulting astrometric parameters for the systems considered here.

Modeling Gaia Epoch Astrometry

The simulation involves modeling along-scan (AL) measurements, which are time-dependent quantities consisting of a time series of along-scan positions with respect to the reference position adopted in the global astrometric solution. The single-star astrometric signal is expressed as wss = (∆α⋆ + µα⋆ t) sin θ + (∆δ + µδt) cos θ + Πϖϖ. p. 2

Where the paper says it

int/gost/ 2. Single-star astrometric signal simulation The epoch astrometry of a single star (SS) represents the change in position due to proper motion (PM) and parallax, projected onto the Gaia along-scan direction: wss = (∆α ⋆ + µα⋆ t) sin θ + (∆δ + µδt) cos θ + Πϖϖ, (1) where ∆α ⋆ and ∆δ denote the offsets from the reference position (α0, δ0), µα⋆ and µδ are the proper motion components, ϖ is the parallax, and Πϖ the corresponding along-scan parallax factors.

Binary astrometric signals are modeled using a hierarchy of complexity, including fixed (VIMF), linear (VIML), acceleration (VIMA), and orbital (VIMO) models for variability-induced movers. p. 2
Where the paper says it

A similar hierarchy exists for variability-induced movers. Wielen (1996) introduced a sequence of models of increasing complexity: fixed (VIMF), linear (VIML), acceleration (VIMA), and orbital (VIMO).

Impact of Binary Motion and VIM Effects on Parameters

Both binary motion and the VIM effect can significantly impact parallax and proper motion measurements, with the magnitude of the effect strongly dependent on the orbital period, photocentric amplitude, and flux ratio of the specific system. Larger time coverage does not necessarily improve parallax precision, in contrast to proper motion, for which precision generally improves with longer baselines. p. 0

Where the paper says it

We show that both binary motion and the VIM effect can significantly impact parallax and proper motion measurements, with the magnitude of the effect strongly dependent on the orbital period, photocentric amplitude, and flux ratio of the specific system. We also show that larger time coverage does not necessarily improve parallax precision, in contrast to proper motion, for which precision generally improves with longer baselines.

Parallaxes retrieved with binary-system fitting are more accurate than those obtained within the single-star framework. p. 0
Where the paper says it

We also show that larger time coverage does not necessarily improve parallax precision, in contrast to proper motion, for which precision generally improves with longer baselines. Instead, parallaxes retrieved with binary-system fitting are more accurate than those obtained within the single-star framework.

Validation and Results

The framework is validated by comparing simulations with existing datasets, such as Gaia BH3, where the simulated along-scan positions matched the published measurements with a χ2 red of 0.65. Proper motion correction for unresolved binaries was shown to be necessary, as subtracting the fitted proper motion from the Gaia DR3 proper motion yields an estimate closer to the true barycentric proper motion. p. 3

Where the paper says it

Subtracting this contribution from the Gaia DR3 proper motion removes the orbital bias and yields an estimate closer to the true barycentric proper motion of the system. We then apply this correction in our simulations to obtain the most realistic representation of the systems considered here.

The study shows that for VIM systems, the parallax can be significantly affected, with scatter observed between different Gaia releases. p. 8
Where the paper says it

This case shows that the VIM effect can introduce significant scatter and strongly affect the resulting parallax. In the case of the remaining five Cepheids, the parallax is still affected by binary motion.

Conclusions and Future Work

The vimbinary tool is presented as a Python package designed to model Gaia DR4-like epoch astrometry, usable both before and after the release. Accounting for these effects with an appropriate model is expected to improve the calibration of the period–luminosity relation with Gaia DR4, both directly and through a better characterisation of the projection factor used in the parallax-of-pulsation method. p. 9

Where the paper says it

The accuracy does not necessarily improve with successive releases, but considering the appropriate model can significantly improve the accuracy of distance determinations. Accounting for these effects with an appropriate model is therefore expected to improve the calibration of the period–luminosity relation with Gaia DR4, both directly and through a better characterisation of the projection factor used in the parallax-of-pulsation method for Cepheid distance determinations (Nardetto et al.

p. 9
Where the paper says it

The accuracy does not necessarily improve with successive releases, but considering the appropriate model can significantly improve the accuracy of distance determinations. Accounting for these effects with an appropriate model is therefore expected to improve the calibration of the period–luminosity relation with Gaia DR4, both directly and through a better characterisation of the projection factor used in the parallax-of-pulsation method for Cepheid distance determinations (Nardetto et al.

The authors note that variability-induced motions are likely to become more important in the case of asymptotic giant branch (AGB) stars due to their large photometric variability. p. 9
Where the paper says it

Together, these improvements will contribute to a more accurate calibration of the local H0 measurement. We note that variability-induced motions are likely to become more important in the case of asymptotic giant branch (AGB) stars due to their large photometric variability.

Additionally, these photocentre-limited observations could be complemented with ground-based observations to resolve the system and break the flux–mass degeneracy. p. 9
Where the paper says it

We will address this question in a second paper (Decin et al. In addition, these photocentre-limited observations could be complemented with ground-based observations to resolve the system, measure the component separation, and break the flux– mass degeneracy, allowing the determination of the component masses.

The authors conclude that for such systems with small photocentric orbits, binary motion does not introduce a significant bias in the parallax. p. 8
Where the paper says it

This is also supported by the agreement between the parameters of AK For derived from the Gaia NSS solution and those from the Gaia main catalogue. We therefore conclude that, for such systems with small photocentric orbits, binary motion does not introduce a significant bias in the parallax.

The scatter in parallax for Cepheid systems remains significant across different releases, and individual case analysis is more suitable for achieving the best precision. p. 9
Where the paper says it

The final parallax value depends on many parameters of the system and their relationships with each other. In order to achieve the best parallax precision for binary Cepheids, it is therefore more suitable to consider each case individually.

Acknowledgements

This work received funding from ANR Unlock-pfactor (ANR-23-CE31-0009). This research is supported by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program (grant agreement No. p. 0

Where the paper says it

González1, 2 1 LIRA, Observatoire de Paris, Université PSL, Sorbonne Université, Université Paris Cité, CY Cergy Paris Université, CNRS, 5 place Jules Janssen, 92195 Meudon, France 2 French-Chilean Laboratory for Astronomy, IRL 3386, CNRS and U. de Chile, Casilla 36-D, Santiago, Chile 3 Instituto de Alta Investigación, Universidad de Tarapacá, Casilla 7D, Arica, Chile 4 Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, 3001, Leuven, Belgium 5 Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, Nice, France 6 UNIDIA, Observatoire de Paris, Université PSL, CNRS, 92190 Meudon, France 7 European Southern Observatory Headquarters, Karl-Schwarzschild-Str.

951549 - UniverScale). PK acknowledges support from the Polish-French Marie Skłodowska-Curie and Pierre Curie Science Prize awarded by the Foundation for Polish Science. p. 0
Where the paper says it

de Chile, Casilla 36-D, Santiago, Chile 3 Instituto de Alta Investigación, Universidad de Tarapacá, Casilla 7D, Arica, Chile 4 Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, 3001, Leuven, Belgium 5 Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, Nice, France 6 UNIDIA, Observatoire de Paris, Université PSL, CNRS, 92190 Meudon, France 7 European Southern Observatory Headquarters, Karl-Schwarzschild-Str. 2, 85748 Garching, Germany 8 Smithsonian Astrophysical Observatory, MS 4, 60 Garden Street, Cambridge, MA 02138, USA 9 European Space Agency (ESA), ESA Office, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA 10 MAUCA – Master track in Astrophysics, Université Côte d’Azur & Observatoire de la Côte d’Azur, Parc Valrose, 06100 Nice, France Received 12 June 2026, Accepted 4 November 2026 ABSTRACT Context.

LD acknowledges support through the European Space Agency (ESA) Research Fellowship in Space Science. p. 0
Where the paper says it

de Chile, Casilla 36-D, Santiago, Chile 3 Instituto de Alta Investigación, Universidad de Tarapacá, Casilla 7D, Arica, Chile 4 Institute of Astronomy, KU Leuven, Celestijnenlaan 200D, 3001, Leuven, Belgium 5 Université Côte d’Azur, Observatoire de la Côte d’Azur, CNRS, Laboratoire Lagrange, Nice, France 6 UNIDIA, Observatoire de Paris, Université PSL, CNRS, 92190 Meudon, France 7 European Southern Observatory Headquarters, Karl-Schwarzschild-Str. 2, 85748 Garching, Germany 8 Smithsonian Astrophysical Observatory, MS 4, 60 Garden Street, Cambridge, MA 02138, USA 9 European Space Agency (ESA), ESA Office, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA 10 MAUCA – Master track in Astrophysics, Université Côte d’Azur & Observatoire de la Côte d’Azur, Parc Valrose, 06100 Nice, France Received 12 June 2026, Accepted 4 November 2026 ABSTRACT Context.

This work has made use of data from the European Space Agency (ESA) mission Gaia processed by the Gaia Data Processing and Analysis Consortium (DPAC). p. 0
Where the paper says it

2, 85748 Garching, Germany 8 Smithsonian Astrophysical Observatory, MS 4, 60 Garden Street, Cambridge, MA 02138, USA 9 European Space Agency (ESA), ESA Office, Space Telescope Science Institute, 3700 San Martin Drive, Baltimore, MD 21218, USA 10 MAUCA – Master track in Astrophysics, Université Côte d’Azur & Observatoire de la Côte d’Azur, Parc Valrose, 06100 Nice, France Received 12 June 2026, Accepted 4 November 2026 ABSTRACT Context. The fourth data release of the Gaia mission (DR4) will, for the first time, provide epoch astrometry.

Funding for the DPAC has been provided by national institutions. p. 0
Where the paper says it

Introduction The Gaia third data release (DR3, Gaia Collaboration et al. p. 0

Where the paper says it

Astrometry – Methods: data analysis – Stars: binaries: close – Stars: variables: Cepheids 1. Introduction The Gaia third data release (DR3, Gaia Collaboration et al.

p. 0
Where the paper says it

Astrometry – Methods: data analysis – Stars: binaries: close – Stars: variables: Cepheids 1. Introduction The Gaia third data release (DR3, Gaia Collaboration et al.

p. 0
Where the paper says it

Astrometry – Methods: data analysis – Stars: binaries: close – Stars: variables: Cepheids 1. Introduction The Gaia third data release (DR3, Gaia Collaboration et al.

2016, 2023b) provided 800 000 binary solutions (Gaia Collaboration et al.

The anonymous referee is thanked for their careful review and constructive comments.

References

Béguin, E., Chiavassa, A., Ahmad, A., Freytag, B., & Uttenthaler, S. 2024, A&A, 690, A125. p. 1

Where the paper says it

2022; Breuval et al. 2024; H0DN Collaboration et al.

Breuval, L. p. 1
Where the paper says it

Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al. 2022; Breuval et al.

, Riess, A. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

G., Casertano, S., et al. 2024, ApJ, 973, 30. p. 1
Where the paper says it

2022; Breuval et al. 2024; H0DN Collaboration et al.

Brown, A. G. A. 2025, Gaia DR4 Processing Status and Contents Preview. p. 1
Where the paper says it

We expect this situation to improve with the upcoming DR4 release (expected on December 2nd of 2026, Brown 2025), which will provide access to epoch astrometry. As these data are not publicly available at the time of writing, in this work, we introduce the vimbinary tool to simulate Gaia DR4-like epoch astrometry, with particular emphasis on VIM effects.

Castelli, F. p. 4
Where the paper says it

As Gallenne et al. (2023) report flux ratios in the K band, we extrapolate these values to the G band using Castelli-Kurucz model atmospheres (Castelli & Kurucz 2003) assuming solar metallicity for the stellar parameters listed in Gallenne et al.

& Kurucz, R. p. 4
Where the paper says it

As Gallenne et al. (2023) report flux ratios in the K band, we extrapolate these values to the G band using Castelli-Kurucz model atmospheres (Castelli & Kurucz 2003) assuming solar metallicity for the stellar parameters listed in Gallenne et al.

L. 2003, in IAU Symposium, Vol. p. 4
Where the paper says it

As Gallenne et al. (2023) report flux ratios in the K band, we extrapolate these values to the G band using Castelli-Kurucz model atmospheres (Castelli & Kurucz 2003) assuming solar metallicity for the stellar parameters listed in Gallenne et al.

210, Modelling of Stellar Atmospheres, ed. p. 4
Where the paper says it

As Gallenne et al. (2023) report flux ratios in the K band, we extrapolate these values to the G band using Castelli-Kurucz model atmospheres (Castelli & Kurucz 2003) assuming solar metallicity for the stellar parameters listed in Gallenne et al.

N. Piskunov, W. W. Weiss, & D. F Gray. Delisle, J. p. 8
Where the paper says it

We simulate the DR3 along-scan observations of these four targets over 1000 bootstrap realisations and fit SS and BS models to them. The BS model fitting is performed with the kepmodel Python package (Delisle & Ségransan 2022).

-B., & Ségransan, D. p. 8
Where the paper says it

We simulate the DR3 along-scan observations of these four targets over 1000 bootstrap realisations and fit SS and BS models to them. The BS model fitting is performed with the kepmodel Python package (Delisle & Ségransan 2022).

, 2022, A&A, 667, A172. p. 1
Where the paper says it

Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al. 2022; Breuval et al.

El-Badry, K. p. 2
Where the paper says it

We accounted for this loss by randomly rejecting 10% of all predicted transits together with an additional 10% of per-CCD points, consistent with real time-sampling. Gaussian Gaia-like measurement noise was then added to the simulated epoch astrometry using the magnitude-dependent empirical CCD along-scan uncertainty prescription described by El-Badry et al.

, Lam, C., Holl, B., et al. 2024, The Open Journal of Astrophysics, 7, 100. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

Evans, N. R. 1995, ApJ, 445, 393. p. 5
Where the paper says it

Because the Gaia G band has a similar mean wavelength to the V band, we use this approximation in the simulations. V1334 Cyg is a first-overtone Cepheid with a B7V companion (Evans 1995).

Evans, N. R., Berdnikov, L., Lauer, J., et al. 2015, AJ, 150, 13. p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

Evans, N. R., Gallenne, A. p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

p. 1
Where the paper says it

No Cepheids were treated as VIMs in DR3, likely reflecting the limited number of VIM solutions in the catalogue (869 in total) and the fact that Cepheids are generally distant objects, so their apparent orbital motion often remains small. As a result, comparisons of parallaxes for binary Cepheids between Gaia DR3 values and those derived from the orbital method, which combines relative astrometry and radial velocity measurements, reveal discrepancies (Gallenne et al.

, Kervella, P. p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

p. 1
Where the paper says it

Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al. 2015; Sana 2017; Kervella et al.

, et al. 2025a, ApJ, 986, 96. p. 5
Where the paper says it

However, no ultraviolet spectra were available to isolate the companion lines, preventing a full determination of system parameters and leaving a degeneracy between masses and distance. The authors estimated a distance of d = 986 ± 48 pc (ϖ = 1.

Evans, N. R., Gallenne, A., Kervella, P., et al. 2024b, ApJ, 972, 145. Everall, A. p. 2
Where the paper says it

The individual AL measurements are assumed to be independent and uncorrelated, following the same simplifying assumption adopted by the astrometric global iterative solution (AGIS, Lindegren et al. 2012), as well as by Everall et al.

, Boubert, D., Koposov, S. E., Smith, L., & Holl, B. p. 2
Where the paper says it

Gaussian Gaia-like measurement noise was then added to the simulated epoch astrometry using the magnitude-dependent empirical CCD along-scan uncertainty prescription described by El-Badry et al. (2024), based on the median DR3 AL uncertainties (see also Holl et al.

2021a MNRAS, 502, 1908. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

Eyer, L., Audard, M., Holl, B., et al. 2023a A&A, 674, A13. p. 0
Where the paper says it

2016, 2023b) provided 800 000 binary solutions (Gaia Collaboration et al. 2023a), which represents a significant increase in the total number of known binaries.

Fulton, B. J., & Petigura, E. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

A. 2018 AJ, 156, 264. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

Gaia Collaboration et al. 2023a A&A, 674, A34. p. 0
Where the paper says it

2016, 2023b) provided 800 000 binary solutions (Gaia Collaboration et al. 2023a), which represents a significant increase in the total number of known binaries.

Gaia Collaboration et al. 2024a A&A, 686, L2. Gaia Collaboration et al. 2013b A&A, 595, A1. Gallenne, A., Evans, N. p. 1
Where the paper says it

At a given time, Gaia observes a single photocentric position determined by both the orbital phase and the pulsation phase of the Cepheid. Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al.

p. 1
Where the paper says it

At a given time, Gaia observes a single photocentric position determined by both the orbital phase and the pulsation phase of the Cepheid. Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al.

p. 1
Where the paper says it

At a given time, Gaia observes a single photocentric position determined by both the orbital phase and the pulsation phase of the Cepheid. Cepheids are prime VIM candidates because they exhibit a high fraction of binary or multiple systems (Evans et al.

R., Kervella, P., et al. 2025a A&A, 693, A111. Gallenne, A., Kervella, P., Borgniet, S., et al. 2019a A&A, 622, A164. p. 6
Where the paper says it

AW Per is a triple system (or possibly quadruple, with an additional companion resolved by Gaia; Kervella et al. 2019a) with a main orbital period of 38 years.

Gallenne, A., Kervella, P., Evans, N. R., et al. 2018b ApJ, 867, 121. Groenewegen, M. p. 1
Where the paper says it

(2024), the derived orbital period coincides with the 33-month time span of DR3, which raises concerns about the robustness of the solution. Nevertheless, Groenewegen (2013) derived an orbital period of 1113.

A. T. 2013 A&A, 550, A70. p. 1
Where the paper says it

(2024), the derived orbital period coincides with the 33-month time span of DR3, which raises concerns about the robustness of the solution. Nevertheless, Groenewegen (2013) derived an orbital period of 1113.

H0DN Collaboration, Casertano, S. p. 1
Where the paper says it

2022; Breuval et al. 2024; H0DN Collaboration et al.

, Anand, G., et al. 2026 A&A, 708, A166. p. 0
Where the paper says it

Astronomy & Astrophysics manuscript no. aa61416-26 ©ESO 2026 November 13, 2026 Astrometric modeling of unresolved variable binary systems I.

Halbwachs, J. p. 0
Where the paper says it

Not only is it expected to contain a drastically increased number of binary solutions compared to DR3, but the community will also have access to epoch astrometry for the first time. This will allow users to conduct more targeted studies and to use complementary data, moving beyond the binary detection performed in DR3 (Halbwachs et al.

-L., Pourbaix, D., Arenou, F., et al. 2023 A&A, 674, A9. p. 0
Where the paper says it

5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al. 2023), the majority of ⋆ Corresponding author: ekaterina.

Holl, B., Sozzetti, A., Sahlmann, J., et al. 2023 a A&A, 674, A10. p. 0
Where the paper says it

5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al. 2023), the majority of ⋆ Corresponding author: ekaterina.

Kervella, P., Arenou, F., & Thévenin, F. 2022 A&A, 657, A7. p. 1
Where the paper says it

Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al. 2022; Breuval et al.

Kervella, P., Gallenne, A., Evans, N. R., et al. 2019b A&A, 623, A116. p. 1
Where the paper says it

2015; Sana 2017; Kervella et al. 2019b), and their pulsational variability is continuous and stable over time, making them particularly suitable for this method (Wielen 1996).

Leavitt, H. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

S. 1908 Annals of Harvard College Observatory, vol. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

60, pp. 87-108. Leavitt, H. S., & Pickering, E. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

C., 1912 Harvard College Observatory Circular, 173, 1. p. 1
Where the paper says it

Moreover, observations of Cepheids with this approach are especially important as Cepheids are a cornerstone of the cosmic distance ladder. Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al.

Lindegren, L. p. 2
Where the paper says it

Right: the corresponding astrometric positions projected onto the Gaia along-scan reference frame. precision, effectively making each observation one-dimensional (Lindegren et al.

p. 2
Where the paper says it

Right: the corresponding astrometric positions projected onto the Gaia along-scan reference frame. precision, effectively making each observation one-dimensional (Lindegren et al.

, Bastian, U., Biermann, M., et al. 2021 A&A, 649, A4. p. 2
Where the paper says it

2012), as well as by Everall et al. (2021) and El-Badry et al.

Lindegren, L., Lammers, U., Hobbs, D., et al. 2012 A&A, 538, A78. p. 2
Where the paper says it

The individual AL measurements are assumed to be independent and uncorrelated, following the same simplifying assumption adopted by the astrometric global iterative solution (AGIS, Lindegren et al. 2012), as well as by Everall et al.

Moe, M. & Di Stefano, R. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

2017 ApJS, 230, 15. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

Nardetto, N. p. 9
Where the paper says it

The accuracy does not necessarily improve with successive releases, but considering the appropriate model can significantly improve the accuracy of distance determinations. Accounting for these effects with an appropriate model is therefore expected to improve the calibration of the period–luminosity relation with Gaia DR4, both directly and through a better characterisation of the projection factor used in the parallax-of-pulsation method for Cepheid distance determinations (Nardetto et al.

, Gieren, W., Storm, J. p. 5
Where the paper says it

As in the case of V350 Sgr, no companion spectral lines were detected, preventing independent distance determinations. Instead, the authors adopted distances from the period–luminosity relation (Storm et al.

, et al. 2023 A&A, 671, A14. p. 0
Where the paper says it

5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al. 2023), the majority of ⋆ Corresponding author: ekaterina.

Neilson, H. p. 1
Where the paper says it

In the Gaia DR4 era, we expect to obtain precise parallaxes for a large sample of Cepheids to calibrate this relation. In addition, Cepheid masses suffer from discrepancies between values derived from evolutionary models and those inferred from pulsation calculations (Neilson et al.

R., Cantiello, M., & Langer, N. 2011 A&A, 529, L9. p. 5
Where the paper says it

Instead, the authors adopted distances from the period–luminosity relation (Storm et al. 2011): U Aql: 592 ± 19 pc (ϖ = 1.

Offner, S. S. R., Moe, M., Kratter, K. M., et al. 2023 in Astronomical Society of the Pacific Conference Series, Vol. p. 1
Where the paper says it

Gaia along-scan astrometry The concept of along-scan (AL) measurements follows directly from the Gaia observing strategy. Rather than measuring positions in right ascension (α) and declination (δ) at each epoch, the satellite continuously scans the sky and records a time series of measurements in its own reference frame: the along-scan abscissa wAL (hereafter simply w) and the across-scan coordinate wAC.

534. Pecaut, M. J., & Mamajek, E. p. 5
Where the paper says it

To estimate companion magnitudes, we assume spectral types from Gallenne et al. (2019) (A0V for U Aql B and B6V for S Mus B) and use Table 5 of Pecaut & Mamajek (2013) to derive Vband magnitudes at the assumed distances: 9.

E., 2013 ApJS, 208, 9. p. 1
Where the paper says it

(2024), the derived orbital period coincides with the 33-month time span of DR3, which raises concerns about the robustness of the solution. Nevertheless, Groenewegen (2013) derived an orbital period of 1113.

Riess, A. G., Yuan, W., Macri, L. M., et al. 2022 ApJ, 934, L7. p. 1
Where the paper says it

Their period–luminosity (PL) relation (Leavitt 1908; Leavitt & Pickering 1912) provides one of the primary routes to determine the Hubble constant (H0) (Riess et al. 2022; Breuval et al.

Sana, H. 2017 in IAU Symposium, Vol. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

329 The Lives and Death-Throes of Massive Stars. p. 0
Where the paper says it

Nevertheless, the whole catalogue contains about 1. 5 billion sources with at least a five-parameter (single-star) solution; given the expected binary fraction of at least 50% for Sun-like stars, rising to nearly 100% for the most massive stars (Moe & Di Stefano 2017; Fulton & Petigura 2018; Offner et al.

Shetye, S. S., Viviani, G., Anderson, R. I., et al. 2024 A&A, 690, A284. p. 1
Where the paper says it

2022; Breuval et al. 2024; H0DN Collaboration et al.

Smolec, R., Singh Rathour, R., & Hocdé, V. 2026 arXiv e-prints arXiv:2605. p. 0
Where the paper says it

Here, we focus on the treatment of binary systems in epoch astrometry in preparation for the DR4. Article number, page 1 arXiv:2611.

14128. Storm, J., Gieren, W., Fouqué, P., et al. 2011 A&A, 534, A94. p. 5
Where the paper says it

Instead, the authors adopted distances from the period–luminosity relation (Storm et al. 2011): U Aql: 592 ± 19 pc (ϖ = 1.

Szabados, L. p. 1
Where the paper says it

One of them, Gaia DR3 5256789350372915328, previously unknown as a binary in the literature, was modelled using a constant-acceleration solution, in which the source motion is approximated by a linear change in proper motion over time. The second, RX Cam, already suspected to host a companion (Szabados 1992), was fitted with a full Keplerian solution.

1992 Information Bulletin on Variable Stars, 3812, 1. p. 1
Where the paper says it

One of them, Gaia DR3 5256789350372915328, previously unknown as a binary in the literature, was modelled using a constant-acceleration solution, in which the source motion is approximated by a linear change in proper motion over time. The second, RX Cam, already suspected to host a companion (Szabados 1992), was fitted with a full Keplerian solution.

Wielen, R. 1996 A&A, 314, 679. p. 1
Where the paper says it

As the flux of the variable component changes, its contribution to the total flux varies, which shifts the position of the photocentre back and forth between the two components (see Fig. This effect was first studied by Wielen (1996) in the context of the Hipparcos mission, with possible extension to Gaia.

Appendix A: VIM Models

The VIM fixed (VIMF) model extends the single-star model with two additional parameters, Dα and Dδ representing the link between astrometric positions and flux of the system. The VIM linear (VIML) and accelerated (VIMA) models approximate linear or accelerated motion of components in the centre-of-mass frame. p. 2

Where the paper says it

Wielen (1996) introduced a sequence of models of increasing complexity: fixed (VIMF), linear (VIML), acceleration (VIMA), and orbital (VIMO). The simpler models approximate the photocentre motion when the orbital signal is only partially resolved, while the VIMO model provides a full Keplerian description of the photocentre displacement.

The VIMO model expresses the orbital contribution using Thiele–Innes constants, where A links astrometric variations to flux variations. p. 3
Where the paper says it

This yields the skyprojected photocentre coordinates α ⋆ orb(t) and δorb(t), which we then use in Eq. Thus, the flux variations translate directly into changes in the photocentre position, and hence photometric and astrometric data can be fitted simultaneously to constrain the variabilityinduced motion (see App.

Appendix B: Fitting Procedure and Uncertainty Estimation

The fitting procedure for linear models involves a weighted least squares solution using the design matrix A and covariance matrix C−1. The VIMA model is non-linear, requiring two scalar parameters k and s to model the dependence of acceleration terms. p. 2

Where the paper says it

Wielen (1996) introduced a sequence of models of increasing complexity: fixed (VIMF), linear (VIML), acceleration (VIMA), and orbital (VIMO). The simpler models approximate the photocentre motion when the orbital signal is only partially resolved, while the VIMO model provides a full Keplerian description of the photocentre displacement.

Parameter uncertainties are estimated using a bootstrap method involving random rejection of predicted transits and per-CCD measurements. p. 2
Where the paper says it

In practice, about 20 percent of the data points are lost because of outlier rejection, dead time, and other effects. We accounted for this loss by randomly rejecting 10% of all predicted transits together with an additional 10% of per-CCD points, consistent with real time-sampling.

The study shows that for non-pulsating binary systems, the error in the DR3 parallax is not dominated by the presence of a companion. p. 8
Where the paper says it

This results from the small contribution of binary motion to the astrometric signal, due to the small photocentric orbits, which act at a level comparable to the astrometric noise. As a result, the error in the DR3 parallax is not dominated by the presence of a companion.

Summary of Findings

The accuracy does not necessarily improve with successive releases, but considering the appropriate model can significantly improve the accuracy of distance determinations. Accounting for these effects with an appropriate model is expected to improve the calibration of the period–luminosity relation with Gaia DR4, both directly and through a better characterisation of the projection factor used in the parallax-of-pulsation method. The study confirms that for systems with small photocentric orbits, binary motion does not introduce a significant bias in the parallax. p. 8

Where the paper says it

This is also supported by the agreement between the parameters of AK For derived from the Gaia NSS solution and those from the Gaia main catalogue. We therefore conclude that, for such systems with small photocentric orbits, binary motion does not introduce a significant bias in the parallax.

The larger time coverage does not cancel the contribution of the parallax, and general scatter in parallax persists across releases. p. 8
Where the paper says it

Rather, we observe general scatter that does not decrease with time. The larger time coverage does not cancel the contribution of the parallax.

The authors conclude that for binary Cepheids, it is more suitable to consider each case individually to achieve the best parallax precision. p. 9
Where the paper says it

The final parallax value depends on many parameters of the system and their relationships with each other. In order to achieve the best parallax precision for binary Cepheids, it is therefore more suitable to consider each case individually.

Key Results

The proper motion correction applied to the raw DR3 values yields better agreement with those reported by Gaia Collaboration et al. (DR4) with differences of ∆µα⋆ = 0. p. 2

Where the paper says it

5 for DR4). 1 is linear in the astrometric parameters (∆α ⋆ , ∆δ, µα⋆ , µδ, ϖ), allowing for a straightforward weighted least-squares solution (see App.

186±0.184 and ∆µδ = −0. p. 5
Where the paper says it

(2024) with difference of ∆µα⋆ = 0. 184 and ∆µδ = −0.

243±0.169 mas a−1. p. 5
Where the paper says it

184 and ∆µδ = −0. 169 mas a−1 .

The simulated photocentric motion for V350 Sgr shows a strong VIM effect with an amplitude of 2. p. 0
Where the paper says it

We present the new vimbinary tool used to simulate and analyse the epoch astrometry data. We show that both binary motion and the VIM effect can significantly impact parallax and proper motion measurements, with the magnitude of the effect strongly dependent on the orbital period, photocentric amplitude, and flux ratio of the specific system.

23 mas. The scatter in parallax for the Cepheid sample ranges from 3. p. 8
Where the paper says it

The other three Cepheids in our sample do not have sufficiently Fig. Parallax comparison for three Cepheid targets and BH3.

63 to 1.34σ across DR2 to DR5. p. 8
Where the paper says it

The scatter is 3. 34σ from DR2 to DR5, respectively, or, in terms of percentage of the true value, 21.

The VIML model provided both the SS parameters and the VIM-specific ones, showing a difference in retrieved parallax between models. p. 8
Where the paper says it

The VIMA model does not significantly improve the VIML model, as expected, because the observed motion is well approximated by a linear trend given the limited orbital coverage. The VIML model provides both the SS parameters and the VIM-specific ones (see equations in App.

The non-pulsating binary sample showed that the error in DR3 parallax is not dominated by the presence of a companion. p. 8
Where the paper says it

This results from the small contribution of binary motion to the astrometric signal, due to the small photocentric orbits, which act at a level comparable to the astrometric noise. As a result, the error in the DR3 parallax is not dominated by the presence of a companion.

Impact on Proper Motion

The proper motion value becomes more accurate with time for systems with orbital periods shorter than the Gaia time coverage, even when using an SS model. The behaviour of AW Per can be explained by its long orbital period, where the DR4 time span covers only about 15% of the orbit. p. 8

Where the paper says it

separation between the components. The second particularity of this star is its long period: the DR4 time span covers only about 15% of the orbit.

For non-pulsating binaries, the error in proper motion decreases with time. p. 7
Where the paper says it

For the non-pulsating binaries, we performed the same analysis, and the result can be seen in Fig. We observed the same trend, with the error decreasing with time.

The final parallax value depends on many parameters of the system and their relationships with each other, requiring individual consideration for best precision. p. 9
Where the paper says it

The final parallax value depends on many parameters of the system and their relationships with each other. In order to achieve the best parallax precision for binary Cepheids, it is therefore more suitable to consider each case individually.

Future Directions

The authors anticipate that variability-induced motions are likely to become more important in the case of asymptotic giant branch (AGB) stars due to their large photometric variability. Convection-induced variations are expected to strongly affect the astrometric solution for AGB stars. p. 9

Where the paper says it

Convection-induced variations were shown to be non-negligible by Béguin et al. Combined with the large photometric variability amplitudes of these stars, convection is therefore expected to strongly affect the astrometric solution.

These photocentre-limited observations could be complemented with ground-based observations to resolve the system and break the flux–mass degeneracy. p. 9
Where the paper says it

We will address this question in a second paper (Decin et al. In addition, these photocentre-limited observations could be complemented with ground-based observations to resolve the system, measure the component separation, and break the flux– mass degeneracy, allowing the determination of the component masses.

The study suggests that including stellar rotation and improving mass determinations in binary systems can help resolve discrepancies between evolutionary models and pulsation calculations. p. 9
Where the paper says it

In addition, these photocentre-limited observations could be complemented with ground-based observations to resolve the system, measure the component separation, and break the flux– mass degeneracy, allowing the determination of the component masses. This could help resolve the discrepancies between values derived from evolutionary models and those inferred from pulsation calculations for Cepheid masses (current attempts to Article number, page 10

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Robust astrometric parameter estimation for unresolved binary systems can be achieved by implementing a framework to simulate Gaia DR4-like epoch astrometry using a vimbinary tool, allowing users to prepare future DR4 analyses.

  2. The improved AI system can accurately quantify the impact of binary motion and variability-induced motion on parallax and proper motion measurements by simulating data under single-star assumptions commonly adopted in Gaia DR3.

  3. The system can perform proper motion correction for unresolved binaries by simulating the effect of orbital motion, such as subtracting the fitted proper motion from the Gaia DR3 proper motion to yield the true barycentric proper motion of the system.

  4. The AI can accurately model variability-induced motions in Cepheids by defining a parameter dvar(t) which defines the width of the 'VIM zone' (see Fig. 1), a band around the mean photocentric orbit that delimits the maximum and minimum photocentric shifts introduced by the VIM effect.

  5. The system can perform full orbital simulations for unresolved binary systems, modeling binarity through a Keplerian orbit with a semi-major axis given by aph = arel B − β, which is crucial for recovering the full twodimensional motion of the source.

  6. The AI can simulate and validate binary models using specific targets, such as Gaia BH3 or AK For, by comparing simulated along-scan positions with published measurements to ensure validating our binary model implementations.

  7. The system can estimate proper motion corrections for Cepheid binaries based on their orbital period, noting that the accuracy improves with time as Gaia covers an increasing fraction of the orbit.

  8. The AI can assess the effect of binarity and VIM on parallax measurements by comparing results from different models, showing that Accounting for these effects with an appropriate model is therefore expected to improve the calibration of the period–luminosity relation with Gaia DR4.

Abstract

Gaia DR4 will, for the first time, provide epoch astrometry. When analysing these data, binary sources must be treated carefully, as modelling their astrometry as single stars can bias parallax and proper motion estimates. We developed the vimbinary tool, a framework to simulate and interpret epoch astrometry of unresolved binary systems and to evaluate the impact of model assumptions on the resulting astrometric parameters. We simulated Gaia DR4-like epoch astrometry for a sample of 17 well-characterised unresolved binary systems, including both non-pulsating binaries and systems containing one photometric variable component - variability-induced movers (VIMs). We show that both binary motion and the VIM effect can significantly impact parallax and proper motion measurements, with the magnitude of the effect strongly dependent on the orbital period, photocentric amplitude, and flux ratio of the specific system. We also show that larger time coverage does not necessarily improve parallax precision, in contrast to proper motion, for which precision generally improves with longer baselines. Instead, parallaxes retrieved with binary-system fitting are more accurate than those obtained within the single-star framework.

Sources

Related papers