Exocomets of $\beta$ Pictoris I: Exocomet destruction, sodium and disk line variability in 17 years of HARPS observations

B. Prinoth; H.J. Hoeijmakers; K.P. Jaworska

REVIEW 3 major objections 6 minor 45 references

Exocomets of $\beta$ Pictoris I: Exocomet destruction, sodium and disk line variability in 17 years of HARPS observations

T0 review · 3 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Sodium absorption from beta Pictoris exocomets is almost never observed, and two 2019 Ca II events show sudden non-Keplerian acceleration that the paper interprets as comet nuclei breaking apart after periastron.

desk verdict A careful HARPS study that establishes rare, Ca II-linked Na I exocomet absorption, but the 2019 non-Keplerian events are not yet proof of nucleus destruction. read the letter →

arxiv 2505.21625 v1 pith:NTFAMIE2 submitted 2025-05-27 astro-ph.EP

H.J. Hoeijmakers , K.P. Jaworska , B. Prinoth This is my paper

classification astro-ph.EP

keywords exocometsbetaPictorisHARPSspectroscopycalciumH&KlinessodiumDcircumstellardiskvariabilitytelluriccorrectionnon-Keplerianacceleration

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper tries to establish two things from nearly two decades of HARPS spectra of beta Pictoris. First, exocometary sodium absorption is rare: in 198 nights only two clear ~2% events plus one weaker ~1% feature lasting 13 nights appear, and they coincide with the deepest Ca II absorptions at the same redshift, implying sodium becomes visible only when the same bodies evaporate calcium extremely strongly, because most sodium is quickly photo-ionized close to the star. Second, two blue-shifted accelerating Ca II features observed on 11 and 12 December 2019 deviate sharply from Keplerian motion within hours, which the authors interpret as destruction of the comet nuclei shortly after periastron passage, analogous to sungrazing comets like Lovejoy. The paper also argues that the Ca II and Na I disk lines vary on observable timescales, so static reference spectra can misattribute disk variability to exocomets. If right, these results sharpen the use of sodium as a tracer of extreme outgassing and add a concrete, observable fate for stargrazing exocomets.

What carries the argument

The analysis rests on three working parts. An unbiased reference spectrum is built by taking the 80th percentile of flux values per wavelength channel and correcting the truncation bias with a truncated-normal formula, so that time-variable exocomet absorption is not baked into the baseline; telluric water lines near the sodium doublet are removed before any sodium analysis. Exocomet lines are modeled as homogeneous Gaussian clouds with central optical depth $\tau_0$, a fill factor $f$ for stellar-disk coverage, and doublet oscillator-strength ratio fixed at 2, fitted with an auto-differentiable Bayesian sampler that yields posterior velocities, widths, depths, and accelerations. For the 2019 features, the paper computes families of Keplerian orbits consistent with the early linear acceleration and then compares the later velocity curves against those predictions, so the departure from Keplerian motion is the observable that carries the destruction claim.

What would settle it

A re-analysis of the 2019 nights that models the Ca II disk line as time-variable, allowing its depth, width, and velocity to drift on hour timescales, would settle the destruction claim: if the apparent acceleration reversal and line broadening disappear once the disk line is free to vary, they are disk variability rather than comet fragmentation. Separately, detecting sodium absorption in an exocomet event with only moderate Ca II depth would directly refute the claim that Na I becomes detectable only during the most extreme Ca II outgassing.

Watch

Extended reading notes

Core claim

After systematically telluric-correcting 9071 HARPS spectra and calibrating them against an unbiased reference spectrum, the paper finds that exocometary Na I is generally absent: only two nights (March 2008 and March 2009) show clear ~2% absorption in both sodium D lines, plus a ~1% feature persisting ~13 nights in February 2004. In all cases the sodium appears red-shifted at the same radial velocity as exceptionally deep Ca II H&K absorption, so the paper concludes that the Na I is produced by the same evaporating bodies, and that detectable sodium requires the most extreme Ca II outgassing because sodium is otherwise photo-ionized. For the 2019 events, the paper shows two blue-shifted Ca II features on consecutive nights that accelerate roughly linearly at first, then suddenly depart from Keplerian acceleration; one reverses its acceleration within about an hour, and line widths grow simultaneously. The paper hypothesizes this is the final fragmentation of the comet nuclei and blow-out of the Ca II tail shortly after periastron, rather than an egress artifact. It further reports a ~5 km/s blue-shifted Ca II feature persisting from early 2017 to mid-2018 that the classical exocomet model may not explain, and documents Na I disk-line depth variations by more than a factor of two as well as a night where the Ca II disk line narrows and shifts, concluding that the circumstellar disk lines themselves are observably variable.

Load-bearing premise

The load-bearing premise is that the 80th-percentile reference spectrum is a faithful static baseline for the stellar and circumstellar disk lines, so that all residual time-variable absorption can be attributed to exocomets; since the paper itself shows the Na I disk lines vary in depth by more than a factor of two and the Ca II disk line can narrow and shift on a single night, disk variability could contaminate the long-lived 2017-2018 feature and the fitted parameters of the 2019 events.

Editorial extensions

If this is right

Sodium D-line absorption can be used as a specific diagnostic of the most extreme Ca II-producing exocomets; most exocomet events will be invisible to sodium surveys.
Multi-species observations of rare deep events can probe the evaporation and ionization structure of exocomet clouds, though abundance ratios will require physical cloud models rather than homogeneous ones.
The 2019 events imply that stargrazing comets can be destroyed within hours near or after periastron, and that sudden non-Keplerian acceleration with line broadening is a signature of that destruction.
The long-lived 2017-2018 Ca II feature and the variable disk lines mean that some 'exocomet' absorption may actually be circumstellar disk gas, so future surveys need time-variable disk models.
If the 2019 pair are on the same orbit, their different behavior after two hours suggests fragmentation is stochastic, not a predictable orbital effect.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If sodium traces only extreme outgassing, comparing Ca II and Na I line depths across a large sample could provide a quantitative threshold for photo-ionization shielding, effectively mapping where in the coma neutral sodium can survive.
The destruction scenario predicts that some stargrazing exocomets should show a characteristic sequence within a single night: linear acceleration, then a sudden acceleration reversal or jump with line broadening; this sequence could be searched for automatically in other archival time-series of beta Pictoris and of other debris-disk stars.
The year-long 2017-2018 feature, if it is disk gas, would connect exocomet activity to the slowly varying circumstellar disk and might change how the Falling Evaporating Bodies population is separated from the disk in future analyses.
A testable extension: re-fit the 2019 nights with a model that lets the disk line parameters vary with time; if the non-Keplerian acceleration survives, the destruction interpretation is strengthened, and if it does not, the event is disk variability.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Two things to know before you read this one. First, it's a genuinely useful observational paper: the first systematic, telluric-corrected Na I exocomet survey over 198 nights of HARPS data, and it finds that sodium absorption is rare and occurs only during the deepest Ca II events. Second, the eye-catching claim—two 2019 comets showing non-Keplerian acceleration, interpreted as post-periastron nucleus destruction—is plausible but far from proven; it rests on single-Gaussian centroid fits that an unresolved second component could mimic.

The paper does a lot right. It grinds through 9071 spectra, corrects tellurics with molecfit, builds a reference spectrum with documented truncation and Tikhonov smoothing, and fits lines with explicit Bayesian priors (Tables A.1, A.2). The Na I detections at ~1–2% depth are supported by fits with decent constraints. The authors also honestly document disk line variability (Figs A.4, A.5), which is relevant because that variability is the main threat to their reference-spectrum assumption.

Where it gets soft: the 'general absence' of Na I is asserted from visual inspection of night-averaged spectra, with no formal detection-limit map. That makes the rarity claim a statement about what's visible, not a quantitative upper limit. The bigger issue is the fragmentation story. The stress-test gets it right: the centroid of a single Gaussian can shift and broaden if a second, weaker component emerges or changes, without any dynamical acceleration. The authors handled the known blend on night 2 by adding a component, but night 1 is still vulnerable, and the line-width jump in Fig 10 is precisely the signature of a blend. Their argument against tail egress doesn't rule out an unresolved blend. Minor point: the conclusions say '2018' while the abstract and results say '2019'—sloppy.

Who's this for? Exocomet and debris-disk researchers will want to see the Na I survey and the 2019 events. It deserves a serious referee. I'd send it out, with a request for a detection-limit analysis and a more careful treatment of unresolved blends before the destruction claim is accepted. That's a revision, not a reject.

Referee Report

3 major / 6 minor

Summary. This paper presents a systematic analysis of 9071 archival HARPS spectra of Beta Pictoris (198 nights, 2003-2020) aimed at finding new exocomet phenomenology in the Ca II H&K lines and the Na I D doublet. The authors apply molecfit-based telluric correction, construct an 80th-percentile reference spectrum with Tikhonov smoothing, and fit the exocomet and disk absorption with a Bayesian multi-component Gaussian model (JAX/NumPyro) with explicit, tabulated priors. The main results are: (i) exocometary Na I absorption is generally absent, with only two ~2%-deep events (March 2008 and March 2009) and a weaker (~1%) feature persisting over 13 nights in February 2004, all coincident in time and radial velocity with exceptionally deep Ca II absorption; (ii) a long-lived blue-shifted Ca II feature near -5 km/s that appears continuously for about a year (2017-2018), which the authors note may be difficult to explain with the classical exocomet picture; and (iii) two blue-shifted, strongly accelerating Ca II features observed on consecutive nights in December 2019 that depart from Keplerian motion, which the authors hypothesize to be evidence for the destruction of the comet nuclei shortly after periastron passage. The paper also documents significant variability in the Na I and Ca II disk lines that complicates the reference-spectrum approach.

Significance. If the results hold, this is a valuable contribution to exocomet studies. The Na I rarity result constrains photoionization and sublimation models, and the simultaneous Na I + Ca II detections provide a physical link between extreme outgassing and detectable neutral sodium that goes beyond the classical Falling Evaporating Bodies framework. The reported disk-line variability is an important caution for the common percentile-based reference-spectrum methodology, and the long-lived 2017-2018 feature challenges existing models. The fragmentation scenario for the 2019 events, with its analogy to Kreutz-family sungrazers such as comet Lovejoy, is an intriguing and falsifiable hypothesis. The paper's strengths include its unusually large and uniform dataset (9071 spectra over 17 years), the systematic telluric correction, transparent Bayesian fits with explicit priors (Tables A.1, A.2), and the authors' own candid statements about the limitations of the reference-spectrum assumption.

major comments (3)

[Section 3.2, Figs. 8-10] The central novel claim - that the 2019 features depart from Keplerian motion because the comet nuclei fragmented - rests entirely on the fitted centroid v0 of single-Gaussian fits to a blended, time-varying absorption region. The paper itself introduces a second static component near -39 km/s on night 2 (Fig. 9), omits unstable fits at the end of that night, and reports a simultaneous dramatic broadening of the line (Fig. 10). A second unresolved or gradually emerging absorption component could shift and broaden the centroid of a single-Gaussian fit without any dynamical acceleration of the absorbing body. The paper's argument against the tail-egress scenario (Section 3.2) does not test this unresolved-blend degeneracy, particularly for night 1 where only one component was fitted. I request a direct test: refit night 1 with a two-component model (or a time-varying blend) and compare the model evidence, and/or show that single-component residuals remain clean throughout the sequence. Without such a test, the attribution of the non-Keplerian departures to nucleus destruction (abstract and Section 4) is not uniquely supported.
[Sections 2.3, 3.3, Figs. A.4-A.6] The reference spectrum is the baseline against which all time-variable absorption is measured, and the paper itself documents that the disk lines vary substantially: the Na I D2 line depth varies by more than a factor of two (from ~5% to ~13%, Fig. A.4), and the Ca II disk line can be narrower and shifted on a single night (Fig. A.5). The long-lived -5 km/s feature of 2017-2018 (Fig. A.6) sits close to the disk line, exactly where the reference spectrum is least certain, and the authors themselves ask whether it might be 'due to some other component of the circumstellar material' (Section 3.3). The impact of the documented disk-line variability on the parameters of the long-lived feature and on the weaker Na I fits is not quantified. I ask that the reference spectrum be perturbed within its stated 0.5-2% uncertainty (or reconstructed with different percentile thresholds) and that the affected claims be re-derived; this is particularly important because the 2019 features are well separated from the disk line, so the reference-spectrum concern is most acute precisely for the long-lived feature claim, not for the fragmentation claim.
[Section 3.1, Figs. A.2-A.3] The claim of a 'general absence' of exocometary Na I absorption, with only two clear (~2%) events out of 198 nights, is established by visual inspection of night-averaged spectra with an ad hoc '>~2%' threshold; no per-night detection limit is given. Given that telluric Na I is present at the 1-5% level in 71 of 198 nights (Fig. A.3), that the Na I disk lines vary in depth by more than a factor of two (Fig. A.4), and that the weakest fitted exocomet feature has a constrained line depth of only 0.89 +/- 0.13% (Fig. 7), the sensitivity of the search should be quantified before concluding that Na I is generally absent. A map of the minimum detectable line depth per night (or per epoch) would make the central sodium rarity claim falsifiable and would also justify the statement that the Earth's atmospheric sodium is not being mistaken for exocometary absorption in the remaining nights.

minor comments (6)

[Section 4] Section 4 states that the two anomalous events were 'observed in 2018', whereas Section 3.2 and the abstract date them to 11 and 12 December 2019; the year should be corrected to 2019.
[Fig. 6 vs. Section 3.1] The caption of Fig. 6 dates the strongest sodium event to 'March 17 2008', while Section 3.1 and Fig. 5 give 'March 16 2008'; the dates should be reconciled.
[Abstract] The abstract contains a typo, 'blue-shifted exocomes', which should read 'exocomets'.
[Section 4, last paragraph] The final paragraph of the conclusions contains a grammatically broken sentence beginning 'We systematically corrected these spectra using Molecfit and that exocometary sodium absorption is relatively rare'; this should be rephrased.
[Section 3.2] The companion paper is referred to inconsistently as both 'Jaworska et al. in prep.' and 'Jarworska et al. in prep.' within the same section; the spelling should be unified.
[Fig. 10] The vertical axis of Fig. 10 is labeled 'Centroid velocity v0 (km/s)' with values ranging from about 24 to 34 km/s, while the text quotes the features at approximately -32 km/s; please clarify whether the plotted quantity is |v0| or else correct the sign convention.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the derivation chain is self-contained, with the reference-spectrum baseline, Bayesian line fits, and forward orbital model all operating independently of the claims they support.

full rationale

The paper's central claims are not circular. The reference spectrum is an 80th-percentile truncated estimator of the static stellar-plus-disk baseline, constructed from the full dataset, but it is not fitted to the target exocomet features; detections are reported as time-variable residual absorption against that baseline, and the authors explicitly acknowledge that disk-line variability could contaminate long-lived features (e.g., 'whether this feature is due to some other component of the circumstellar material'). The exocomet line parameters are obtained with a Bayesian model with stated priors, including explicit handling of blended components on night 2, so the fitted centroids and widths are not assumed detections. The orbital analysis is a forward calculation: the authors enumerate Keplerian orbits consistent with the observed linear acceleration segment and then compare later data against that family, rather than fitting an orbit to the entire time series and calling the residuals a prediction. The fragmentation hypothesis is presented as an interpretation of the observed non-Keplerian acceleration and line broadening, with the alternative tail-egress scenario explicitly considered and argued against. Self-citations to the authors' earlier work concern data-reduction practices, and the companion paper (Jaworska et al., in prep.) is used only to motivate plausibility of an icy origin, not as the basis for any derived quantity. No fitted parameter is renamed as a prediction, no uniqueness theorem is imported, and no known result is repackaged under new coordinates. The paper is therefore self-contained against external benchmarks for its stated inferences, and the minor year inconsistency between the abstract (2019) and conclusions (2018) is a presentational error, not circularity.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central observational claims rest on a handful of modeling assumptions rather than on derived constants. The main risks are reference-spectrum construction and telluric correction, both acknowledged as imperfect. No new physical entities are postulated.

free parameters (4)

Reference spectrum truncation threshold T = 80th percentile of per-channel flux distribution
Section 2.3: chosen after comparing reference spectra with thresholds from 10th to 98th percentile; this choice sets the baseline against which all exocomet and disk line depths are measured.
Tikhonov regularization strength lambda = 1
Section 2.3, Eq. 3: smoothing of the reference spectrum; value is chosen by hand rather than derived.
Na I fill fraction f = 1 (fixed)
Section 3.1: f and tau0 are degenerate for optically thin lines; f is fixed to 100%, making tau0 a lower limit and the line depth D the meaningful fitted quantity.
Line component count and priors per fit = 3 components for Na I fits; 4 components for Ca II night-2 fits
Section 2.4 and Tables A.1-A.2: the number of Gaussian components and their priors are selected per event by eye, not by an automated model-selection criterion.

assumptions (5)

domain assumption The absorbing exocomet cloud is homogeneous and produces Gaussian line profiles, with the doublet optical depth ratio fixed by oscillator strengths (Eq. 4-6).
Section 2.4: used to translate observed line depths into tau0, f, v0, and sigma. Real clouds may have radial density gradients, which the authors note in Section 3.1.
domain assumption The reference spectrum captures the static stellar and disk-line baseline, so residual time-variable absorption can be attributed to exocomets.
Section 2.3: central to all line-depth measurements. The paper itself shows disk-line variability in Figs. A.4 and A.5, so this assumption is only partially valid.
domain assumption Telluric correction with molecfit using 10 water lines adjacent to Na I leaves no significant residuals that mimic exocomet features.
Section 2.2: identification of rare Na I depends on clean removal of telluric water lines. Section 3.1 documents residual telluric sodium at 1-5%, so residuals are known to exist.
domain assumption Orbital families are computed under two-body Keplerian motion with inclination i = 0 and longitude of ascending node Omega = 180 degrees.
Section 3.2 and Fig. 11: used to judge consistency with Keplerian acceleration and to define the claimed departure from Keplerian motion. Radiation pressure and ion-tail effects add unknown velocity offsets, as the authors note.
domain assumption Uncertainties are photon-noise-limited and equal to the square root of the recorded counts for e2ds spectra.
Section 2.1: used in the likelihood and in the beta scaling factor that absorbs unmodeled variance. Correlated noise beyond the fitted beta is assumed absent.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Exocomets of $\beta$ Pictoris I: Exocomet destruction, sodium and disk line variability in 17 years of HARPS observations." pith.science (2026). https://pith.science/paper/NTFAMIE2

@misc{pith2026250521625,
  author       = {Pith},
  title        = {Pith review of: Exocomets of $\beta$ Pictoris I: Exocomet destruction, sodium and disk line variability in 17 years of HARPS observations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NTFAMIE2}},
  note         = {Machine review of arXiv:2505.21625}
}

abstract

The young $\beta$ Pictoris system has been monitored with high-resolution optical spectrographs for decades. These observations have revealed strongly variable absorption in the Ca II H\&K lines attributed to in-falling cometary bodies. Since 2003, over 9000 HARPS observations of $\beta$ Pictoris have been taken and many of these have not yet been used for exocomet studies. We search these spectra for new exocomet phenomenology enabled by the long time coverage and large volume of this dataset. We systematically carry out telluric correction of the HARPS spectra using molecfit, compare multi-year observations of the Ca II and Na I lines and use a Bayesian fitting algorithm to extract exocomet line parameters. We explore the usage of an unbiased reference spectrum with which to calibrate the continuum, and investigate Keplerian orbital solutions to observed exocomet acceleration. We find a general absence of exocometary sodium line absorption, with only two instances of clear ($\sim 2$ % deep) exocometary sodium out of 198 nights of observation, as well as a weaker ($\sim 1$ %) feature that persists over 13 nights in 2004. We find that these events occur during times of deep Ca II absorption at the same red-shift, implying that strongly Ca II-evaporating exocomets also exhibit detectable levels of Na I, in spite of the vast majority of Na I being rapidly photo-ionised in close proximity to the star. We find long-lived Ca II absorption in 2017 and 2018 that persists on a timescale of a year, which may be difficult to explain with the classical exocomet model. Finally, we investigate two strongly accelerating, blue-shifted exocomes observed in 2019 that show strong and sudden departures from Keplerian motion, suggesting rapid changes to the dynamics of the exocomet cloud. We hypothesize that this is caused by the destruction of the comet nuclei shortly after their periastron passages.

Figures

Figures reproduced from arXiv: 2505.21625 by the authors.

**Figure 2.** Reference spectrum obtained following the methodology described in section 2.3 (red), compared to the simple time averaged spectrum (green) and the HARPS data (black). Smoothing using Tikhonov regularization reduces the noise in the reference spectrum at the 0.5% level (inset). As expected for an unbiased estimator, the reference spectrum tends to the time-averaged spectrum at wavelengths where no exocomet activ… view at source ↗

**Figure 3.** Example of a fit to the Ca II doublet (K-line in the top panel and H-line in the bottom panel) of a blue-shifted exocomet observed in the night of December 10, 2019. The exocomet is modeled as a single component, and the disk line as a blend of two. The narrow fitting region is restricted to the blue side of the disk line to avoid additional unrelated exocomet components. Because the Ca II lines are covered by thre… view at source ↗

Figures from the paper (6 more)

**Figure 4.** Figure 4: Posterior distributions for five relevant model parameters of the fit shown in [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]

**Figure 5.** Figure 5: Telluric-corrected HARPS observations at the sodium doublet, obtained on two nights nearly one year apart, vertically offset for clarity. In both nights, significant disk line absorption is visible, as well as red-shifted lines attributed to a form of exocomet activity…

**Figure 6.** Figure 6: Best-fit models to the Na I D lines (left and middle columns) on three nights of data (top, middle and bottom rows), compared to the Ca II K line (right column). The night of March 17 2008 (middle row) shows the strongest absorption of exocometary sodium of all 198 nig…

**Figure 7.** Figure 7: Posterior distributions of the parameters that describe the exocometary sodium lines on the night of February 17, 2004 shown in [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

**Figure 10.** Figure 10: Centroid velocities (top panel) and Gaussian line widths (bottom panel) of the exocomet features shown in [PITH_FULL_IMAGE:figures/full_fig_p009_10.png]

**Figure 11.** Figure 11: Family of orbital solutions that have a mean radial velocity during the transit between -35 and -27.5 km/s and a mean radial acceleration between 0.9 and 1.3 km/s/h, consistent with the linear part of the exocomet radial velocity observed during the first three hours…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

45 extracted references · 34 canonical work pages

[1]

Aumann , H. H. 1985, Publications of the Astronomical Society of the Pacific, 97, 885

work page 1985
[2]

2017, The Convergence of Markov chain Monte Carlo Methods : From the Metropolis method to Hamiltonian Monte Carlo

Betancourt, M. 2017, The Convergence of Markov chain Monte Carlo Methods : From the Metropolis method to Hamiltonian Monte Carlo

work page 2017
[3]

M., Crawford , I

Beust , H., Lagrange , A. M., Crawford , I. A., et al. 1998, , 338, 1015

work page 1998
[4]

2024, , 683, A89

Beust , H., Milli , J., Morbidelli , A., et al. 2024, , 683, A89

work page 2024
[5]

P., Jankowiak, M., et al

Bingham, E., Chen, J. P., Jankowiak, M., et al. 2018, Pyro: Deep Universal Probabilistic Programming , arXiv:1810.09538

arXiv 2018
[6]

2018, JAX : composable transformations of P ython+ N um P y programs

Bradbury, J., Frostig, R., Hawkins, P., et al. 2018, JAX : composable transformations of P ython+ N um P y programs

work page 2018
[7]

H., Li , A., Bohac , C., et al

Chen , C. H., Li , A., Bohac , C., et al. 2007, , 666, 466

work page 2007
[8]

H., Lu , C

Chen , C. H., Lu , C. X., Worthen , K., et al. 2024, arXiv e-prints, arXiv:2407.04661

arXiv 2024

Show all 45 references

[9]

M., & Vidal-Madjar , A

Ferlet , R., Hobbs , L. M., & Vidal-Madjar , A. 1987, , 185, 267

1987
[10]

M., Lagrange-Henri , A

Hobbs , L. M., Lagrange-Henri , A. M., Ferlet , R., Vidal-Madjar , A., & Welty , D. E. 1988, , 334, L41

1988
[11]

J., Kitzmann , D., Morris , B

Hoeijmakers , H. J., Kitzmann , D., Morris , B. M., et al. 2024, , 685, A139

2024
[12]

J., Seidel , J

Hoeijmakers , H. J., Seidel , J. V., Pino , L., et al. 2020, , 641, A123

2020
[13]

2001, , 372, 616

Karmann , C., Beust , H., & Klinger , J. 2001, , 372, 616

2001
[14]

2015, , 576, A78

Kausch , W., Noll , S., Smette , A., et al. 2015, , 576, A78

2015
[15]

Kennedy , G. M. 2018, , 479, 1997

2018
[16]

2014, , 514, 462

Kiefer , F., Lecavelier des Etangs , A., Boissier , J., et al. 2014, , 514, 462

2014
[17]

2019, , 621, A58

Kiefer , F., Vidal-Madjar , A., Lecavelier des Etangs , A., et al. 2019, , 621, A58

2019
[18]

J., Rodet , L., et al

Lacour , S., Wang , J. J., Rodet , L., et al. 2021, , 654, L2

2021
[19]

M., De Bondt , K., Meunier , N., et al

Lagrange , A. M., De Bondt , K., Meunier , N., et al. 2012, , 542, A18

2012
[20]

M., Ferlet , R., & Vidal-Madjar , A

Lagrange , A. M., Ferlet , R., & Vidal-Madjar , A. 1987, , 173, 289

1987
[21]

M., Kasper , M., Boccaletti , A., et al

Lagrange , A. M., Kasper , M., Boccaletti , A., et al. 2009, , 506, 927

2009
[22]

M., Meunier , N., Rubini , P., et al

Lagrange , A. M., Meunier , N., Rubini , P., et al. 2019, Nature Astronomy, 3, 1135

2019
[23]

M., Beust , H., Ferlet , R., & Vidal-Madjar , A

Lagrange-Henri , A. M., Beust , H., Ferlet , R., & Vidal-Madjar , A. 1989, , 215, L5

1989
[24]

2022, Scientific Reports, 12, 5855

Lecavelier des Etangs , A., Cros , L., H \'e brard , G., et al. 2022, Scientific Reports, 12, 5855

2022
[25]

1999, , 343, 916

Lecavelier Des Etangs , A., Vidal-Madjar , A., & Ferlet , R. 1999, , 343, 916

1999
[26]

X., Chen , C

Lu , C. X., Chen , C. H., Sargent , B. A., et al. 2022, , 933, 54

2022
[27]

& Queloz , D

Mayor , M. & Queloz , D. 1995, , 378, 355

1995
[28]

& Takeuchi , T

Murata , K. & Takeuchi , T. T. 2022, , 74, 1329

2022
[29]

K., Kataza , H., Honda , M., et al

Okamoto , Y. K., Kataza , H., Honda , M., et al. 2004, , 431, 660

2004
[30]

2022, , 660, A49

Pavlenko , Y., Kulyk , I., Shubina , O., et al. 2022, , 660, A49

2022
[31]

2002, The Messenger, 110, 9

Pepe , F., Mayor , M., Rupprecht , G., et al. 2002, The Messenger, 110, 9

2002
[32]

2019, arXiv preprint arXiv:1912.11554

Phan, D., Pradhan, N., & Jankowiak, M. 2019, arXiv preprint arXiv:1912.11554

2019 arXiv
[33]

2006, , 373, 231

Pont , F., Zucker , S., & Queloz , D. 2006, , 373, 231

2006
[34]

J., Kitzmann , D., et al

Prinoth , B., Hoeijmakers , H. J., Kitzmann , D., et al. 2022, Nature Astronomy, 6, 449

2022
[35]

& Lai , D

Rodet , L. & Lai , D. 2024, , 527, 11664

2024
[36]

& Chodas , P

Sekanina , Z. & Chodas , P. W. 2012, , 757, 127

2012
[37]

2015, , 576, A77

Smette , A., Sana , H., Noll , S., et al. 2015, , 576, A77

2015
[38]

Smith , B. A. & Terrile , R. J. 1984, Science, 226, 1421

1984
[39]

Tallis, G. M. 1961, Journal of the Royal Statistical Society: Series B (Methodological), 23, 223

1961
[40]

M., Ferlet , R., Gry , C., & Albert , C

Vidal-Madjar , A., Hobbs , L. M., Ferlet , R., Gry , C., & Albert , C. E. 1986, , 167, 325

1986
[41]

2024, , 684, A210

Vrignaud , T., Lecavelier des Etangs , A., Kiefer , F., et al. 2024, , 684, A210

2024
[42]

A., & Kiefer , F

Vrignaud , T., Lecavelier des Etangs , A., Strom , P. A., & Kiefer , F. 2025, arXiv e-prints, arXiv:2503.17346

2025 arXiv
[43]

N., Holman , M

Winn , J. N., Holman , M. J., Torres , G., et al. 2008, , 683, 1076

2008
[44]

2021, , 645, A23

Zhao , F., Lo Curto , G., Pasquini , L., et al. 2021, , 645, A23

2021
[45]

A., & Kennedy , G

Zieba , S., Zwintz , K., Kenworthy , M. A., & Kennedy , G. M. 2019, , 625, L13

2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Aumann , H. H. 1985, Publications of the Astronomical Society of the Pacific, 97, 885

work page 1985

[2] [2]

2017, The Convergence of Markov chain Monte Carlo Methods : From the Metropolis method to Hamiltonian Monte Carlo

Betancourt, M. 2017, The Convergence of Markov chain Monte Carlo Methods : From the Metropolis method to Hamiltonian Monte Carlo

work page 2017

[3] [3]

M., Crawford , I

Beust , H., Lagrange , A. M., Crawford , I. A., et al. 1998, , 338, 1015

work page 1998

[4] [4]

2024, , 683, A89

Beust , H., Milli , J., Morbidelli , A., et al. 2024, , 683, A89

work page 2024

[5] [5]

P., Jankowiak, M., et al

Bingham, E., Chen, J. P., Jankowiak, M., et al. 2018, Pyro: Deep Universal Probabilistic Programming , arXiv:1810.09538

arXiv 2018

[6] [6]

2018, JAX : composable transformations of P ython+ N um P y programs

Bradbury, J., Frostig, R., Hawkins, P., et al. 2018, JAX : composable transformations of P ython+ N um P y programs

work page 2018

[7] [7]

H., Li , A., Bohac , C., et al

Chen , C. H., Li , A., Bohac , C., et al. 2007, , 666, 466

work page 2007

[8] [8]

H., Lu , C

Chen , C. H., Lu , C. X., Worthen , K., et al. 2024, arXiv e-prints, arXiv:2407.04661

arXiv 2024