pith. sign in

arxiv: 2606.24132 · v1 · pith:C2OEUIPAnew · submitted 2026-06-23 · 🌌 astro-ph.EP · astro-ph.IM· astro-ph.SR

Bayesian analysis of Gaia epoch astrometry and radial velocities with kima

Pith reviewed 2026-06-25 23:20 UTC · model grok-4.3

classification 🌌 astro-ph.EP astro-ph.IMastro-ph.SR
keywords Gaia astrometryexoplanet detectionradial velocitiesorbit fittingBayesian inferencenested samplingbinary stars
0
0 comments X

The pith

kima now includes models for Bayesian fitting of Gaia epoch astrometry alone or with radial velocities.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper adds two new models to the open-source kima code to process epoch astrometry measurements from Gaia. These models can run on astrometry data by itself or combined with radial velocity observations. The authors test the implementation on both real datasets and simulated ones, finding that the recovered parameters and evidences match published or expected values. This matters because the next Gaia data release is projected to yield tens of thousands of new exoplanet detections plus orbital solutions for binaries and black holes. The code relies on diffusive nested sampling to perform parameter estimation and model comparison.

Core claim

Two Gaia-specific models were added to kima; when applied to real and simulated epoch astrometry data the resulting orbital parameters and model evidences are consistent with published or expected values, validating the implementation for Gaia analysis and allowing exploration of scan-angle dependent signals versus true orbits.

What carries the argument

Two added Gaia epoch astrometry models inside the kima codebase that supply the likelihood function for diffusive nested sampling.

If this is right

  • Joint astrometry-plus-radial-velocity fits become feasible for thousands of new Gaia detections.
  • The same framework can supply three-dimensional orbital constraints on binaries and black holes.
  • Model evidence comparisons can help separate genuine orbits from scan-angle artifacts.
  • The open-source code is now available for community use on the expected large volume of Gaia exoplanet candidates.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The approach could be applied to astrometric data from future missions beyond Gaia without major code changes.
  • Combining the kima Gaia model with other surveys might tighten constraints on long-period companions.
  • Further tests on edge-case orbits would clarify the limits of distinguishing scan-angle signals.

Load-bearing premise

That agreement between kima outputs and published values on the chosen test datasets means the new models correctly represent the Gaia likelihood for arbitrary future data.

What would settle it

A fresh Gaia dataset on which kima returns orbital parameters or Bayes factors that differ significantly from those produced by an independent Gaia astrometry pipeline.

Figures

Figures reproduced from arXiv: 2606.24132 by Jean-Baptiste Delisle, Jo\~ao P. Faria, Thomas A. Baycroft.

Figure 1
Figure 1. Figure 1: Phase plot of the maximum-likelihood model for Target 1. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Corner plot of the orbital parameters for Target 1 with the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of the posterior distribution in parallax [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Top-left: posterior density for a second Keplerian in the analysis of Target 2. Top-right: posterior density for a second [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
read the original abstract

Context. The next data release from the Gaia space telescope is expected to result in tens of thousands of newly detected exoplanets, as well as detection and 3D orbital constraints on binary stars and even black holes. Many of these will warrant in-depth analyses of the astrometric data, as well as radial velocity follow-up. This will require dedicated tools to exploit this wealth of data. Aims. We provide open-source software to analyse epoch astrometric data from Gaia, both alone and simultaneously with radial velocities. Methods. We add two models to the open-source orbit-fitting codebase kima, and test these on both real an simulated data. This code uses diffusive nested sampling to explore the parameter space, calculate evidences for model comparison, and perform parameter estimation. Results. We show that the results obtained are consistent with published/expected values, validating the use of kima for analysis of Gaia data. We explore various attributes of kima's Gaia model including using it to potentially distinguish true orbits from scan-angle dependent signals.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper extends the open-source kima orbit-fitting package by adding two models for Gaia epoch astrometry (alone or jointly with radial velocities). It employs diffusive nested sampling to perform parameter estimation and compute Bayesian evidences for model comparison. The authors test the implementation on a set of real and simulated systems and report that the recovered parameters and evidences are consistent with published or expected values, thereby validating the models for Gaia data analysis. They additionally examine the model's capacity to distinguish genuine orbital signals from scan-angle-dependent artifacts.

Significance. If the likelihood implementation is correct, the work supplies a publicly available tool for the detailed analysis of the tens of thousands of exoplanet and binary detections expected from future Gaia data releases. The use of nested sampling to obtain evidences enables rigorous model comparison, which is a methodological strength. The open-source release and joint astrometry-RV capability address a clear community need for reproducible Gaia orbit fitting.

major comments (2)
  1. [Abstract] Abstract and Results: the central validation claim rests on consistency with published/expected values, yet no quantitative metrics (e.g., parameter offsets, residual rms, or evidence ratios) are supplied for the test cases. This leaves open the possibility that an incorrect likelihood (missing cross terms or mishandled scan-angle covariances) could still reproduce the chosen test results.
  2. [Methods] Methods (Gaia model description): the two added models are stated to capture epoch astrometry likelihood including scan-angle effects, but no explicit verification against synthetic edge cases (e.g., high proper-motion objects or epochs with strong along-scan error correlations) is described. Such tests are load-bearing for the claim that the implementation is generally usable.
minor comments (2)
  1. [Abstract] The abstract states 'real an simulated data' (typo for 'and').
  2. [Methods] Notation for the Gaia-specific parameters (e.g., scan-angle projections) should be defined explicitly in the first methods subsection where the models are introduced.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful review and positive evaluation of the work's significance. We address the two major comments point by point below, agreeing that additional quantitative details and tests will strengthen the manuscript. We will incorporate the suggested revisions in the next version.

read point-by-point responses
  1. Referee: [Abstract] Abstract and Results: the central validation claim rests on consistency with published/expected values, yet no quantitative metrics (e.g., parameter offsets, residual rms, or evidence ratios) are supplied for the test cases. This leaves open the possibility that an incorrect likelihood (missing cross terms or mishandled scan-angle covariances) could still reproduce the chosen test results.

    Authors: We agree that the absence of explicit quantitative metrics limits the strength of the validation claim. In the revised manuscript we will add a new table (or expanded results section) reporting parameter offsets, residual RMS values, and evidence ratios for all real and simulated test cases. These additions will allow direct assessment of the likelihood implementation and address the concern about possible missing cross terms or covariance handling. revision: yes

  2. Referee: [Methods] Methods (Gaia model description): the two added models are stated to capture epoch astrometry likelihood including scan-angle effects, but no explicit verification against synthetic edge cases (e.g., high proper-motion objects or epochs with strong along-scan error correlations) is described. Such tests are load-bearing for the claim that the implementation is generally usable.

    Authors: We acknowledge that dedicated verification on synthetic edge cases is important for demonstrating general usability. In the revised Methods section we will include new simulations and results for high proper-motion objects and epochs exhibiting strong along-scan error correlations, confirming that the likelihood correctly handles these regimes. revision: yes

Circularity Check

0 steps flagged

No circularity; validation against external published values

full rationale

The paper adds two Gaia astrometry models to the existing kima codebase and validates them solely by showing that posterior parameters and model evidences match published or expected values on independent real and simulated datasets. This is a direct external benchmark comparison with no self-referential fitting, no parameter renamed as prediction, and no load-bearing self-citation chain. The derivation chain consists of standard Bayesian sampling plus explicit likelihood implementations whose correctness is tested externally rather than assumed by construction. No step reduces to its own inputs.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The work relies on the pre-existing kima framework and diffusive nested sampling; no new free parameters, physical entities, or ad-hoc axioms are introduced in the abstract.

axioms (1)
  • domain assumption Diffusive nested sampling correctly computes Bayesian evidences and posterior distributions for the added Gaia orbit models
    The paper uses this method for model comparison and parameter estimation without additional justification in the abstract.

pith-pipeline@v0.9.1-grok · 5724 in / 1173 out tokens · 26095 ms · 2026-06-25T23:20:18.261366+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

63 extracted references · 52 canonical work pages · 2 internal anchors

  1. [1]

    The Journal of Open Source Software , author =

    kima:. The Journal of Open Source Software , author =. 2018 , note =. doi:10.21105/joss.00487 , abstract =

  2. [2]

    2022 , note =

    Monthly Notices of the Royal Astronomical Society , author =. 2022 , note =. doi:10.1093/mnras/stac113 , abstract =

  3. [3]

    and Babusiaux, C

    Leclerc, A. and Babusiaux, C. and Arenou, F. and van Leeuwen, F. and Bonnefoy, M. and Delfosse, X. and Forveille, T. and Bouquin, J.-B. Le and Rodet, L. , month = sep, year =. Combining

  4. [4]

    Journal of the American Statistical Association , author =

    Kass, Robert E. and Raftery, Adrian E. , month = jun, year =. Bayes. Journal of the American Statistical Association , publisher =. doi:10.1080/01621459.1995.10476572 , abstract =

  5. [5]

    Statistics and Computing , author =

    Diffusive nested sampling , volume =. Statistics and Computing , author =. 2011 , file =. doi:10.1007/s11222-010-9198-8 , abstract =

  6. [6]

    Monthly Notices of the Royal Astronomical Society , author =

    Improving circumbinary planet detections by fitting their binary's apsidal precession , volume =. Monthly Notices of the Royal Astronomical Society , author =. 2023 , note =. doi:10.1093/mnras/stad607 , abstract =

  7. [8]

    The Astrophysical Journal Supplement Series , author =

    The. The Astrophysical Journal Supplement Series , author =. 2021 , note =. doi:10.3847/1538-4365/abf93c , abstract =

  8. [9]

    2018 , keywords =

    Journal of Statistical Software , author =. 2018 , keywords =. doi:10.18637/jss.v086.i07 , abstract =

  9. [10]

    Journal of Hydrology , author =

    A generalized probability density function for double-bounded random processes , volume =. Journal of Hydrology , author =. 1980 , note =. doi:10.1016/0022-1694(80)90036-0 , abstract =

  10. [11]

    , volume =

    Parametrizing the exoplanet eccentricity distribution with the beta distribution. , volume =. Monthly Notices of the Royal Astronomical Society , author =. 2013 , note =. doi:10.1093/mnrasl/slt075 , abstract =

  11. [12]

    2023a, MNRAS, 518, 1057, doi: 10.1093/mnras/stac3140

    A. Monthly Notices of the Royal Astronomical Society , author =. 2023 , note =. doi:10.1093/mnras/stac3140 , abstract =

  12. [13]

    The Astrophysical Journal , author =

    How. The Astrophysical Journal , author =. 2010 , note =. doi:10.1088/0004-637X/709/1/168 , abstract =

  13. [14]

    Monthly Notices of the Royal Astronomical Society , author =

    New evidence about. Monthly Notices of the Royal Astronomical Society , author =. 2023 , note =. doi:10.1093/mnras/stad2794 , abstract =

  14. [15]

    2008, Contemporary Physics, 49, 71, doi: 10.1080/00107510802066753

    Bayes in the sky:. Contemporary Physics , author =. 2008 , note =. doi:10.1080/00107510802066753 , abstract =

  15. [16]

    and Fischer, D

    Lovis, C. and Fischer, D. , month = dec, year =. Radial

  16. [17]

    and Lindegren, Lennart , month = dec, year =

    Perryman, Michael and Hartman, Joel and Bakos, Gáspár A. and Lindegren, Lennart , month = dec, year =. Astrometric. The Astrophysical Journal , publisher =. doi:10.1088/0004-637X/797/1/14 , abstract =

  17. [18]

    and Berger, James O

    Jefferys, William H. and Berger, James O. , year =. Ockham's. American Scientist , publisher =

  18. [19]

    Bayesian

    Gregory, Phil , year =. Bayesian. doi:10.1017/CBO9780511791277 , abstract =

  19. [20]

    Astrometry , isbn =

    Perryman, Michael , year =. Astrometry , isbn =. The. doi:10.1017/9781108304160.004 , urldate =

  20. [21]

    Astronomy and Astrophysics , author =

    Gaia. Astronomy and Astrophysics , author =. 2023 , note =. doi:10.1051/0004-6361/202244161 , abstract =

  21. [22]

    and Ségransan, D

    Sahlmann, J. and Ségransan, D. and Queloz, D. and Udry, S. and Santos, N. C. and Marmier, M. and Mayor, M. and Naef, D. and Pepe, F. and Zucker, S. , month = jan, year =. Search for brown-dwarf companions of stars , volume =. Astronomy and Astrophysics , publisher =. doi:10.1051/0004-6361/201015427 , abstract =

  22. [23]

    and Fabricius, C

    Holl, B. and Fabricius, C. and Portell, J. and Lindegren, L. and Panuzzo, P. and Bernet, M. and Castañeda, J. and Jevardat de Fombelle, G. and Audard, M. and Ducourant, C. and Harrison, D. L. and Evans, D. W. and Busso, G. and Sozzetti, A. and Gosset, E. and Arenou, F. and De Angeli, F. and Riello, M. and Eyer, L. and Rimoldini, L. and Gavras, P. and Mowl...

  23. [24]

    Summary of the content and survey properties

    Gaia. Astronomy and Astrophysics , publisher =. 2023 , note =. doi:10.1051/0004-6361/202243940 , abstract =

  24. [25]

    The Astrophysical Journal , author =

    The. The Astrophysical Journal , publisher =. 2022 , note =. doi:10.3847/1538-4357/ac7c74 , abstract =

  25. [26]

    Computing in Science and Engineering , keywords =

    Matplotlib:. Computing in Science & Engineering , author =. 2007 , keywords =. doi:10.1109/MCSE.2007.55 , abstract =

  26. [27]

    Prusti et al

    The. Astronomy and Astrophysics , publisher =. 2016 , note =. doi:10.1051/0004-6361/201629272 , abstract =

  27. [28]

    Faria, J. P. and Delisle, J.-B. and Ségransan, D. , month = dec, year =. Inferring planet occurrence rates from radial velocities , volume =. Astronomy and Astrophysics , publisher =. doi:10.1051/0004-6361/202557774 , abstract =

  28. [29]

    and Faria, J

    Figueira, P. and Faria, J. P. and Silva, A. M. and Castro-González, A. and Gomes da Silva, J. and Sousa, S. G. and Bossini, D. and Zapatero-Osorio, M. R. and Balsalobre-Ruza, O. and Lillo-Box, J. and Tabernero, H. M. and Adibekyan, V. and Allart, R. and Benatti, S. and Bouchy, F. and Cabral, A. and Cristiani, S. and Dumusque, X. and González-Hernández, J....

  29. [30]

    Monthly Notices of the Royal Astronomical Society, Advance Access , author =

    Predicting. Monthly Notices of the Royal Astronomical Society, Advance Access , author =. 2026 , file =. doi:10.1093/mnras/stag450 , abstract =

  30. [31]

    Astronomy and Astrophysics , publisher =

    Discovery of a dormant 33 solar-mass black hole in pre-release. Astronomy and Astrophysics , publisher =. 2024 , note =. doi:10.1051/0004-6361/202449763 , abstract =

  31. [32]

    , month = nov, year =

    Breivik, Katelyn and Chatterjee, Sourav and Larson, Shane L. , month = nov, year =. Revealing. The Astrophysical Journal , publisher =. doi:10.3847/2041-8213/aa97d5 , abstract =

  32. [33]

    and Agrawal, Poojan and Delfavero, Vera and Chatterjee, Sourav , month = apr, year =

    Nagarajan, Pranav and El-Badry, Kareem and Chawla, Chirag and Di Carlo, Ugo Niccolò and Breivik, Katelyn and Rodriguez, Carl L. and Agrawal, Poojan and Delfavero, Vera and Chatterjee, Sourav , month = apr, year =. Realistic. Publications of the Astronomical Society of the Pacific , publisher =. doi:10.1088/1538-3873/adc839 , abstract =

  33. [34]

    and Marcussen, Marcus L

    Stefánsson, Gudmundur and Mahadevan, Suvrath and Winn, Joshua N. and Marcussen, Marcus L. and Kanodia, Shubham and Albrecht, Simon and Fitzmaurice, Evan and Mikulskytė, Onė and Cañas, Caleb I. and Espinoza-Retamal, Juan I. and Zwart, Yiri and Krolikowski, Daniel M. and Hotnisky, Andrew and Robertson, Paul and Alvarado-Montes, Jaime A. and Bender, Chad F. ...

  34. [35]

    Halbwachs, Jean-Louis and Pourbaix, Dimitri and Arenou, Frédéric and Galluccio, Laurent and Guillout, Patrick and Bauchet, Nathalie and Marchal, Olivier and Sadowski, Gilles and Teyssier, David , month = jun, year =. Gaia. Astronomy and Astrophysics , publisher =. doi:10.1051/0004-6361/202243969 , abstract =

  35. [36]

    10.1051/0004-6361/202243782

    Gaia. Astronomy and Astrophysics , publisher =. 2023 , note =. doi:10.1051/0004-6361/202243782 , abstract =

  36. [37]

    and Ségransan, D

    Delisle, J.-B. and Ségransan, D. , month = nov, year =. Analytical determination of orbital elements using. Astronomy and Astrophysics , publisher =. doi:10.1051/0004-6361/202244018 , abstract =

  37. [39]

    2023, The Astronomical Journal, 166, 164, doi: 10.3847/1538-3881/acf5cc

    Thompson, William and Lawrence, Jensen and Blakely, Dori and Marois, Christian and Wang, Jason and Giordano, Mosé and Brandt, Timothy and Johnstone, Doug and Ruffio, Jean-Baptiste and Ammons, S. Mark and Crotts, Katie A. and Do Ó, Clarissa R. and Gonzales, Eileen C. and Rice, Malena , month = sep, year =. Octofitter:. The Astronomical Journal , publisher ...

  38. [40]

    Thiele, Thorvald Nicolai , month = jan, year =. Neue. Astronomische Nachrichten , publisher =

  39. [41]

    Circular of the Union Observatory Johannesburg , author =

    Orbital. Circular of the Union Observatory Johannesburg , author =. 1926 , note =

  40. [42]

    and Hobbs, D

    Ranalli, P. and Hobbs, D. and Lindegren, L. , month = jun, year =. Astrometry and exoplanets in the. Astronomy and Astrophysics , publisher =. doi:10.1051/0004-6361/201730921 , abstract =

  41. [43]

    , month = jan, year =

    Lammers, Caleb and Winn, Joshua N. , month = jan, year =. On the. The Astronomical Journal , publisher =. doi:10.3847/1538-3881/ae21de , abstract =

  42. [44]

    , month = may, year =

    Hilditch, Ronald W. , month = may, year =. An

  43. [45]

    and Lu, Tiger and Armitage, Philip J

    Yahalomi, Daniel A. and Lu, Tiger and Armitage, Philip J. and Bedell, Megan and Casey, Andrew R. and Price-Whelan, Adrian M. and Rice, Malena , month = mar, year =. The. The Astrophysical Journal , publisher =. doi:10.3847/2041-8213/ae42cd , abstract =

  44. [46]

    doi:10.5281/zenodo.3515526 , url =

    Johannes Sahlmann , title =. doi:10.5281/zenodo.3515526 , url =

  45. [47]

    , year = 2014, month = nov, journal =

    Brewer, Brendon J. , year = 2014, month = nov, journal =. Inference for. 1411.3921 , abstract =

  46. [48]

    Brewer, B. J. and Donovan, C. P. , year = 2015, month = mar, journal =. Fast. doi:10.1093/mnras/stv199 , urldate =

  47. [49]

    , year = 2018, month = jun, journal =

    Butkevich, Alexey G. , year = 2018, month = jun, journal =. Astrometric Detectability of Systems with Unseen Companions: Effects of the. doi:10.1093/mnras/sty686 , urldate =

  48. [50]

    Faria, J. P. and Haywood, R. D. and Brewer, B. J. and Figueira, P. and Oshagh, M. and Santerne, A. and Santos, N. C. , year = 2016, month = apr, journal =. Uncovering the Planets and Stellar Activity of. doi:10.1051/0004-6361/201527899 , urldate =

  49. [51]

    Faria, J. P. and Adibekyan, V. and. Decoding the Radial Velocity Variations of. A&A , volume =. doi:10.1051/0004-6361/201936389 , urldate =

  50. [52]

    Faria, J. P. and Su. A Candidate Short-Period Sub-. A&A , volume =. doi:10.1051/0004-6361/202142337 , urldate =

  51. [53]

    MNRAS , volume =

    Lessons Learned from the Detection of Wide Companions by Radial Velocity and Astrometry , author =. MNRAS , volume =. doi:10.1093/mnras/staf689 , urldate =

  52. [54]

    Frensch, Y. G. C. and Lo Curto, G. and Bouchy, F. and Mayor, M. and H. The. Astron. Astrophys. , volume =. doi:10.1051/0004-6361/202346203 , urldate =

  53. [55]

    and Millman, K

    Harris, Charles R. and Millman, K. Jarrod and. Array Programming with. Nature , volume =

  54. [56]

    Anna and Collier Cameron, A

    John, A. Anna and Collier Cameron, A. and Faria, J. P. and Mortier, A. and Wilson, T. G. and Malavolta, L. and Buchhave, L. A. and Dumusque, X. and. Sub-m s-1 Upper Limits from a Deep. MNRAS , volume =. doi:10.1093/mnras/stad2381 , urldate =

  55. [57]

    Stellar and Substellar Companions from

    Kervella, Pierre and Arenou, Fr. Stellar and Substellar Companions from. A&A , volume =. doi:10.1051/0004-6361/202142146 , urldate =

  56. [58]

    A&A , volume =

    Planetary System. A&A , volume =. doi:10.1051/0004-6361/202038922 , urldate =

  57. [59]

    Nanobind: Tiny and Efficient

  58. [60]

    Cold giant discoveries from a joint radial-velocity and astrometry framework

    Cold Giant Discoveries from a Joint Radial-Velocity and Astrometry Framework , author =. A&A , volume =. doi:10.1051/0004-6361/202557940 , urldate =. 2603.11202 , primaryclass =

  59. [61]

    and Petralia, A

    Piccinini, G. and Petralia, A. and Sozzetti, A. and Benatti, S. and Gandolfi, D. and Micela, G. , year = 2026, month = mar, journal =. True Masses Using Radial Velocity Data with. doi:10.1051/0004-6361/202556735 , urldate =

  60. [62]

    Thiele, Thorvald Nicolai , year = 1883, month = jan, journal =. Neue

  61. [63]

    Orbital. Circ. Union Obs. Johannesbg. , volume =

  62. [64]

    , year = 2021, month = jul, journal =

    Venner, Alexander and Vanderburg, Andrew and Pearce, Logan A. , year = 2021, month = jul, journal =. True. doi:10.3847/1538-3881/abf932 , urldate =

  63. [65]

    Paul and Feng, Fabo and Shectman, Stephen A

    Wu, Yiyun and Xiao, Guang-Yao and Butler, R. Paul and Feng, Fabo and Shectman, Stephen A. and Crane, Jeffrey D. and Teske, Johanna K. and Wang, Sharon X. and Beletsky, Yuri and Burt, Jennifer A. and Daylan, Tansu and Diaz, Matias and Dragomir, Diana and Flowers, Erin and Jenkins, Sydney and Kanodia, Shubham and Nandakumar, Sangeetha and Rice, Malena and S...