Pith. sign in

REVIEW 5 major objections 4 minor 89 references

Hot exozodiacal dust around Fomalhaut: The MATISSE perspective

T0 review · 5 major / 4 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 MATISSE L-band data hint at a hot dust ring 0.11 au from Fomalhaut.

desk verdict A careful MATISSE first-look at Fomalhaut whose own error bars support a candidate excess, not the 'second L-band HEZD detection' the abstract claims; the fitting-comparison result is the real contribution. read the letter →

arxiv 2506.02826 v1 pith:62XK6MNH submitted 2025-06-03 astro-ph.EP astro-ph.GA

Kevin Ollmann , Florian Kirchschlager , Thomas A. Stuber , Katsiaryna Tsishchankava , Alexis Matter , Steve Ertel , Tim D. Pearce , Alexander V. Krivov
show 1 more author
Sebastian Wolf
This is my paper
classification astro-ph.EPastro-ph.GA
keywords hotexozodiacaldustFomalhautlong-baselineinterferometryMATISSEVLTIdebrisdisksvisibilityfittingcircumstellar
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

Using the first LM-band MATISSE/VLTI observations of Fomalhaut, the paper asks whether hot exozodiacal dust radiates close to the star and what its properties are. The L-band data show a marginal visibility deficit relative to the bare photosphere, which the authors interpret as circumstellar radiation and attribute to hot dust; this would be only the second L-band detection of such dust, after $\kappa$ Tuc. Fitting an optically thin ring to the data yields an inner radius of $0.11\,\mathrm{au}$, an outer radius of $0.12\,\mathrm{au}$, dust grains around $0.53\,\mu\mathrm{m}$, and a total mass of $3.25\times 10^{-10}\,M_\oplus$, consistent with earlier Fomalhaut studies. The detection is not significant at the $3\sigma$ level in most fits, and an unresolved stellar companion can reproduce the data almost as well, so the case rests on a marginal deficit rather than a secure detection.

What carries the argument

The load-bearing object is the visibility deficit: the ratio between the measured squared visibility of Fomalhaut and the expected visibility of its limb-darkened photosphere, which drops by about one to two percent near $3.06\,\mu\mathrm{m}$. The authors model this deficit either as an optically thin, geometrically narrow dust ring or as a spherical shell, compute the combined stellar-plus-dust visibility with the Van Cittert-Zernike theorem from brightness maps generated by the DMS debris-disk simulation tool, and fit the resulting visibilities directly to the MATISSE data in a self-consistent one-step approach. A two-step approach, bootstrapping, and neural-network emulators are used as cross-checks; the companion test replaces the dust with a point-like secondary on a circular orbit. The machinery matters because the inferred dust parameters depend on which geometric model is assumed and on the fitting path, while the companion test shows how easily a marginal point-source signal can masquerade as dust.

What would settle it

A single deeper MATISSE L-band observation of Fomalhaut with roughly twice the integration time and an independent calibrator pair would settle the central claim: if the 1-2% visibility deficit near $3.06\,\mu\mathrm{m}$ does not reproduce, or if closure phases reveal an off-axis point-source signature whose contrast and orbital motion match a $0.3\,\mathrm{au}$, 2-3% companion, then the hot-dust ring parameters are not a detection. Alternatively, observing at an epoch a few months later and seeing the deficit's signature move with orbital phase would favor the companion.

Watch

Extended reading notes

Core claim

The paper's central claim is that the MATISSE L-band visibilities of Fomalhaut contain a marginal excess of circumstellar emission over the limb-darkened photosphere, most plausibly thermal radiation from hot exozodiacal dust, and that this excess is the second of its kind seen in the L band. On the authors' best-fit narrow-ring model the dust occupies a very thin annulus from $0.11\,\mathrm{au}$ to $0.12\,\mathrm{au}$, with a narrow grain-size distribution centered near $0.53\,\mu\mathrm{m}$ and a total mass of $3.25\times 10^{-10}\,M_\oplus$, with amorphous carbon as the preferred species. The same data are consistent with a spherical shell, and the parameters agree with constraints from earlier VINCI and KIN observations; adding those older data does not tighten the allowed ranges. The authors stress that the geometry assumption matters more than the fitting method for the derived dust-to-star flux ratio, and that neither the dust morphology nor the possibility of a stellar companion can be decisively established from these data alone.

Load-bearing premise

The whole dust interpretation assumes that the small visibility deficit is emission from an optically thin dust ring or shell; the data fit an unresolved 2-3% companion almost as well, and no calibration or instrumental systematic can be excluded, so if the deficit is not dust, the derived ring radius, grain size, and mass do not describe anything real.

Editorial extensions

If this is right

  • Fomalhaut becomes the second star, after $\kappa$ Tuc, with a published L-band hot-exozodiacal-dust signal, and the first such target observed with MATISSE in the LM bands.
  • The derived ring location at $0.11$-$0.12\,\mathrm{au}$ places the dust at or near the carbon sublimation radius, with dust temperatures near 920-1995 K, close to the 2000 K sublimation limit for amorphous carbon.
  • Combining MATISSE with the older VINCI and KIN fluxes does not improve the parameter constraints, so the L/M-band data are consistent with, but not more informative than, earlier K- and N-band measurements.
  • The dust-to-star flux ratio depends more strongly on the assumed ring-versus-Gaussian geometry than on the fitting technique, which means published HEZD properties for other stars may depend on that modeling choice.
  • A stellar companion with about 2-3% contrast near $0.3\,\mathrm{au}$ fits the visibilities and closure phases essentially as well as the dust ring, so the companion hypothesis remains open.

Reading between the lines

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

  • Editorial inference: if the marginal deficit is real dust, the confinement of roughly $0.5\,\mu\mathrm{m}$ grains to a very thin annulus near the sublimation rim argues for a replenishment or trapping mechanism acting on short timescales, since such grains should be blown out or vaporized quickly around an A3 V star.
  • Editorial inference: applying the one-step, self-consistent fitting approach to HEZD systems previously analyzed only with the two-step approach could shift their inferred flux ratios and dust parameters by amounts comparable to the spread seen here, so some published HEZD masses and locations may need revision.
  • Editorial inference: a targeted closure-phase search on longer baselines or at multiple epochs should discriminate between the dust-ring and companion interpretations, because a companion's orbital motion would change the closure-phase signature while a symmetric ring would not.
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.

Referee Report

5 major / 4 minor

Summary. The paper presents the first VLTI/MATISSE L- and M-band interferometric observations of Fomalhaut, obtained with the AT array at medium spectral resolution. The authors fit the calibrated visibilities with limb-darkened photosphere plus a geometrically thin dust ring or spherical shell, using a one-step approach, the traditional two-step approach, a bootstrap resampling method, and neural networks, and they also fit a binary-companion model. They report a marginal visibility deficit in the L band, with best-fit ring parameters of inner radius 0.11 au, outer radius 0.12 au, grain radius 0.53 micron, and total dust mass 3.25e-10 Earth masses, and they characterize this as the second L-band HEZD detection after kappa Tuc. The paper also compares the resulting dust-to-star flux ratios across methods and finds that the assumed geometric model affects the derived flux ratio more than the fitting approach does.

Significance. The manuscript is useful as the first MATISSE L/M-band dataset on Fomalhaut and as a systematic comparison of fitting strategies for HEZD interferometric data. The authors are transparent about many limitations: they state that the two-step analysis yields no 3-sigma detection, that a stellar companion cannot be excluded, and that the AIC/BIC differences are below the significance threshold. The paper also confirms consistency with earlier VINCI/KIN constraints. However, the central 'second L-band HEZD detection' claim rests on a single bootstrap point at one wavelength and a model choice that the data cannot discriminate, so the significance of the result as a dust detection is overstated. The strengths are the careful data-reduction description, the multi-method approach, and the explicit acknowledgment of degeneracies; the weakness is that the abstract and conclusions go beyond what the reported statistics support.

major comments (5)
  1. [Section 4.5, Table 2] The only 3-sigma result in the paper is the bootstrap Gaussian-model flux ratio at lambda = 3.06 micron, obtained by resampling N = 6 baselines as fully correlated units (Appendix A). With such a small, discrete resampling ensemble, the 16-84% quantile interval is not a calibrated significance test for systematic-dominated errors, and no correction is made for the twelve wavelengths and the multiple models (Gaussian vs ring) examined. At the same wavelength, the two-step analysis in Table 2 gives f2/Delta-f2 approximately 0.85. The abstract and Conclusion 1 therefore overstate the evidence when they call this the second L-band HEZD detection; the data support a marginal visibility deficit, and the phrase 'detection' should be replaced by 'candidate' unless the bootstrap significance is properly calibrated for multiple testing and for the small number of independent baselines.
  2. [Section 4.2] The companion analysis shows that a binary with contrast of about 2-3% and semi-major axis around 0.3 au reproduces both the visibilities and closure phases essentially as well as the dust models, with neither the semi-major axis nor the contrast constrained. Because the excess is therefore morphologically unresolved, the ring parameters quoted in the abstract and in Table 3 (Rin = 0.11 au, Rout = 0.12 au, a = 0.53 micron, M = 3.25e-10 Earth masses) are derived under a geometric assumption that the data do not discriminate. While the main text acknowledges this in Section 4.2, the abstract and conclusions present the HEZD interpretation more firmly; please ensure that the abstract and conclusions state explicitly that the companion interpretation is equally consistent and that the dust parameters are model-dependent.
  3. [Section 4.3, Table 4] For the MATISSE data alone, Table 4 reports no error intervals for any fitted parameter, and the AIC/BIC differences (ring AIC = -390 vs photosphere AIC = -387) are below the Delta approximately 5 threshold stated by the authors. The specific best-fit values are therefore an unconstrained minimum of the model grid rather than a measurement of the HEZD properties. The quoted values in the abstract should be labeled as model-dependent best fits with no MATISSE-only error bars, or the abstract should state the absence of constraints explicitly.
  4. [Eq. (7)] Equation (7) does not appear to be the correct visibility of a uniform annulus. With k = 2*pi*Rout*B/lambda, the second Bessel-function argument should be k*(Rin/Rout), i.e., 2*pi*Rin*B/lambda, whereas the printed formula contains k*(Rout/Rin). As written, the equation describes an aperture with an inverted radius ratio and would bias the bootstrap ring flux ratios in Table 2 and Section 4.5; please correct the formula and rerun the affected fits.
  5. [Sections 2 and 4.1] The reported best-fit reduced chi-squared values are well below unity (e.g., 0.07-0.66 for the two-step fits and 0.23 for the one-step ring model). This indicates that the visibility errors are overestimated or strongly correlated. Because both the detection significance and the parameter error intervals (Eq. 6) are computed from these chi-squared values, the quantitative error bars and significance statements should be revisited, for example by rescaling errors to yield chi-squared_red near unity or by using a covariance-aware statistic.
minor comments (4)
  1. [Abstract and Section 4.5] The abstract and the conclusions call the result the 'second detection of HEZD emission in the L band,' while the body repeatedly labels the detection 'marginal' and the AIC/BIC analysis non-significant; please harmonize the wording so the abstract and conclusions reflect the same level of confidence as the body.
  2. [Figure 1] The figure caption lists baselines such as 'K0-D0 (31 m)' without defining these station labels in the text; a short sentence identifying the AT station naming would help the reader.
  3. [Section 4.3] The sentence 'The significance criterion in the context of these criteria only applies to a difference of about five' is grammatically unclear; please rephrase and cite a standard threshold for the AIC/BIC difference.
  4. [Section 4.7 and Table 2] Table 2 includes the M-band wavelengths but the bootstrap columns are left blank there; the text should state once more that bootstrapping was applied only to the L band, since a reader may otherwise wonder whether the M-band blanks are missing values.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the HEZD parameters and flux ratios are explicitly fitted to the MATISSE visibilities, and the paper transparently reports the marginal significance and degeneracies.

full rationale

The paper's derivation chain is model fitting, not prediction. The HEZD parameters (Rin, Rout, a, M) and all dust-to-star flux ratios (f1, f2, fRing, fGaussian, fNN) are obtained by fitting model visibilities to the same MATISSE data; Sect. 3.2 states that the authors 'fit model parameters of a geometrically and optically thin dust ring ... directly to the measured visibilities.' Sect. 4.7 explicitly says that in the one-step approach the flux ratio is 'subsequently derived based on the best-fit parameter values,' rather than presented as an independent prediction. This transparency means no fitted input is renamed as a prediction. The paper is also unusually explicit about the evidential weaknesses: Sect. 4.1 finds no 3-sigma detection in the two-step analysis, Sect. 4.2 shows an equally good binary-companion fit with unconstrained parameters, Sect. 4.3 reports AIC/BIC differences below the threshold of about five and states that MATISSE-only error intervals do not exist, Sect. 4.4 finds the double-ring model neither implied nor excluded, and Sect. 4.7 reports that the one-step flux errors have no defined limits. These passages weaken the detection claim but are honesty about fitting degeneracy, not circular reasoning. The cited prior results from overlapping author groups (bootstrapping recipes, DMS code, the kappa Tuc L-band detection) involve different instruments, targets, or independently published data, and no load-bearing claim reduces to a self-citation or to an assumption already containing the target result. The central outputs are fitted quantities, clearly labeled as such, so there is no by-construction equivalence between inputs and outputs.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. The central deduction is a standard visibility-fit inversion with five fitted dust parameters. The strongest hidden assumptions are the geometric model family, the neglect of temporal variability when combining epochs, and the photosphere and calibration reference models. The neural network approach (Appendix B) adds the assumption that a face-on narrow ring with a specific density exponent spans the parameter space.

free parameters (2)
  • Fitted dust ring parameters (Rin, Rout, a, i, M) = Rin=0.11 au, Rout=0.12 au, a=0.53 micron, i=90 deg, M=3.25e-10 M_Earth
    These five parameters are varied and fitted to the MATISSE visibilities in the one-step approach (Sect. 3.3, Table 1). The derived dust-to-star flux ratio stated in the abstract and conclusions is computed from these fitted parameters.
  • Fitted dust-to-star flux ratios f2 (two-step), fRing/fGaussian (bootstrap), fNN (neural network) = f2 up to 1.84%, fRing up to 2.42%, fGaussian up to 2.02%, fNN ~0.90% at 3.06 micron
    In the two-step, bootstrap, and neural network approaches the flux ratio is either a direct fit parameter or derived from a fit, not predicted from an independent model.
assumptions (4)
  • domain assumption The observed visibility deficit is caused by a geometrically thin, optically thin dust ring or spherical shell of amorphous carbon, silicate, or graphite grains.
    Sect. 3.3 defines the HEZD model family. The companion hypothesis in Sect. 4.2 shows this is not the only viable interpretation of the same data.
  • domain assumption The dust-to-star flux ratio is constant across the observing epochs used to combine MATISSE (2022) with VINCI and KIN data from earlier epochs.
    Stated explicitly in Sect. 4.3: 'we neglect the possible temporal variability of the HEZD radiation.' This is load-bearing for the combined SED fit in Fig. 5 and Table 4.
  • domain assumption The stellar photosphere model (limb-darkened disk, diameter 2.22 mas, limb-darkening coefficient 0.13) is correct at the percent level in the L and M bands.
    Eq. 1 in Sect. 3.1. A wrong stellar diameter or limb darkening would shift the reference visibility against which the deficit is measured.
  • domain assumption The calibration error is dominated by the reported per-channel errors and the 2% broadband error is negligible.
    Sect. 2 states the broadband calibration error was neglected because it appeared lower than per-channel data errors. Given the deficit is at the 1% level, this assumption directly affects the detection significance.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hot exozodiacal dust around Fomalhaut: The MATISSE perspective." pith.science (2026). https://pith.science/paper/62XK6MNH

@misc{pith2026250602826,
  author       = {Pith},
  title        = {Pith review of: Hot exozodiacal dust around Fomalhaut: The MATISSE perspective},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/62XK6MNH}},
  note         = {Machine review of arXiv:2506.02826}
}
abstract

Excess over the stellar photospheric emission of main-sequence stars has been found in interferometric near-infrared observations, and is attributed to the presence of hot exozodiacal dust (HEZD). As part of our effort to detect and characterize HEZD around the nearby A3 V star Fomalhaut, we carried out the first interferometric observations with the MATISSE instrument at the VLTI in the photometric bands L and M for the Fomalhaut system. Assuming a dust distribution either as a narrow ring or spherical shell for modeling the HEZD, we aim to constrain the HEZD parameters by generating visibilities and fitting them to the MATISSE data using different approaches. The L band data provide a marginal detection of circumstellar radiation, potentially caused by the presence of HEZD, which is only the second detection of HEZD emission in the L band. An analysis of the data with different fitting approaches showed that the best-fit values for the HEZD parameters are consistent with those of previous Fomalhaut observations, which underlines the functionality of MATISSE. Assuming a dust ring, it would have an inner ring radius of \(0.11\ \mathrm{au}\), an outer ring radius of \(0.12\ \mathrm{au}\), a narrow dust grain size distribution around a dust grain radius constrained by \(0.53\ \mu\mathrm{m}\), and a total dust mass of \(3.25\times 10^{-10}\ \rm{M}_{\oplus}\). Finally, the results indicate that the choice of the geometric model has a more significant impact on the derived dust-to-star flux ratio than the specific fitting approach applied. Since different dust-to-star flux ratios can result from the applied fitting approaches, this also has an impact on the parameter values of the HEZD around Fomalhaut and most likely for other HEZD systems. Moreover, further NIR and MIR data are required for a more comprehensive description of the emission originating in the vicinity of Fomalhaut.

Figures

Figures reproduced from arXiv: 2506.02826 by the authors.

Figure 1
Figure 1. Calibrated and merged measurement results at the different base￾lines with dependence on the spatial frequency (Mega-λ). The dots de￾note the L band data and the squares the M band data, respectively (see Sect. 2 for details). considered the L and M band data since Fomalhaut is too faint for absolute visibility measurements in the N band with the ATs. We further restricted the data to the wavelength range of ∼ 3.0 −… view at source ↗
Figure 3
Figure 3. Closure phases for the four triangles of the observation (see Sect. 2 and Sect. 4.1 for details). 3. Modeling approach We outline two of our applied fitting approaches (see Ap￾pendix A and Appendix B for the outlines of the other applied approaches) in Sect. 3.1 and Sect. 3.2. Finally, we describe the HEZD model that we applied to constrain the properties of the HEZD around Fomalhaut in Sect. 3.3. 3.1. Two-step appr… view at source ↗
Figure 4
Figure 4. Visibility of the limb-darkened photosphere, best-fit ring model from the one-step approach, uniform circumstellar radiation from the two-step approach (Gaussian response function of MATISSE), and the calibrated MATISSE data dependent on the baseline at three represen￾tative wavelengths. Top: 3.06 µm (L band). Mid: 3.54 µm (L band). Bottom: 4.92 µm (M band, see Sect. 4.1 and Sect. 4.3 for details). phases. The analy… view at source ↗
Figures from the paper (5 more)
Figure 5
Figure 5. Figure 5: Spectral energy distribution of the HEZD inferred from K, L, M, and N band observations with a best-fit parameter configuration for the ring model (see [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Constructed χ 2 red-map for a relative width of Rout/Rin = 1.01 varying the inner ring radii and dust grain radii using neural networks (see Sect. 4.6 for details) [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Constructed χ 2 red-map for the best-fit parameters and varying in￾clinations (i) and position angles (PA) of the ring using neural networks (see Sect. 4.6 for details). mass (range of 10−10 − 10−8 M⊕) that results in the lowest possi￾ble value of χ 2 red, by applying …
Figure 8
Figure 8. Figure 8: Dust-to-star flux ratio with dependence on the wavelength with a best-fit parameter configuration applying the different fitting approaches from the previous subsections (see Sect. 4.7 for details). 4.7. Flux ratios for the fitting approaches Because different dust-to-…
Figure 9
Figure 9. Figure 9: Visibility of the limb-darkened photosphere. Visualized are all applied fitting approaches with best-fit parameter values (see Sect. 4) and calibrated MATISSE data with dependence on the baseline at a wavelength of 3.06 µm (see Sect. 5 for details). the inclination, ha…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

89 extracted references · 77 canonical work pages

  1. [1]

    2016, Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2016, 265–283

    Abadi, M., Barham, P., Chen, J., et al. 2016, Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation, OSDI 2016, 265–283

    2016

  2. [2]

    2006, A&A, 452, 237

    Absil, O., di Folco, E., Mérand, A., et al. 2006, A&A, 452, 237

    2006

  3. [3]

    B., et al

    Absil, O., Mennesson, B., Le Bouquin, J. B., et al. 2009, ApJ, 704, 150

    2009

  4. [4]

    B., Berger, J

    Absil, O., Le Bouquin, J. B., Berger, J. P., et al. 2011, A&A, 535, A68

    2011

  5. [5]

    2013, A&A, 555, A104

    Absil, O., Defrère, D., Coudé du Foresto, V., et al. 2013, A&A, 555, A104

    2013

  6. [6]

    2021, A&A, 651, A45

    Absil, O., Marion, L., Ertel, S., et al. 2021, A&A, 651, A45

    2021

  7. [7]

    D., Herter, T

    Adams, J. D., Herter, T. L., Lau, R. M., Trinh, C., & Hankins, M., 2018, ApJ, 862, 161

    2018

  8. [8]

    2012, A&A, 540, A125

    Acke, B., Min, M., Dominik, C., et al. 2012, A&A, 540, A125

    2012

Show all 89 references
  1. [9]

    Akaike, H., 1974, IEEE Trans. Autom. Control, 19, 6

    1974

  2. [10]

    L., Ciardi, D

    Akeson, R. L., Ciardi, D. R., Millan-Gabet, R., et al. 2009, ApJ, 691, 1896

    2009

  3. [11]

    Backman, D. E. & Paresce, F., 1993, in Protostars and Planets III, ed. E. H. Levy & J. I. Lunine, 1253

    1993

  4. [12]

    C., & Thébault, P., 2012, A&A, 548, A104

    Bonsor, A., Augereau, J. C., & Thébault, P., 2012, A&A, 548, A104

    2012

  5. [13]

    N., Augereau, J

    Bonsor, A., Raymond, S. N., Augereau, J. C., & Ormel, C. W., 2014, MNRAS, 441, 2380

    2014

  6. [14]

    Born, M., Wolf, E., 1980, Principles of Optics: Electromagnetic Theory of Propagation, Interference and Diffraction of Light, 6th edition, Pergamon Press, p. 416, eq. 25

    1980

  7. [15]

    2016, A&A, 589, A112

    Chelli, A., Duvert, G., Bourgès, L., et al. 2016, A&A, 589, A112

    2016

  8. [16]

    & Others

    Chollet, F. & Others. 2015, Keras, https://keras.io

    2015

  9. [17]

    M., Wizinowich, P

    Colavita, M. M., Wizinowich, P. L., Akeson, R. L., et al. 2013, PASP, 125, 1226

    2013

  10. [18]

    C., et al

    Defrère, D., Absil, O., Augereau, J. C., et al. 2011, A&A, 534, A5

    2011

  11. [19]

    2004, A&A, 426, 2

    Di Folco, E., Th\' e venin, F., Kervella, P., et al. 2004, A&A, 426, 2

    2004

  12. [20]

    C., et al

    Di Folco, E., Absil, O., Augereau, J. C., et al. 2007, A&A, 475, 243

    2007

  13. [21]

    Draine, B. T. & Malhotra, S., 1993, ApJ, 414, 632

    1993

  14. [22]

    A., 2017, A&A, 597, A8

    Duvert, G., Young, J., & Hummel, C. A., 2017, A&A, 597, A8

    2017

  15. [23]

    Efron, B., 1979, Ann. Statist. 7(1): 1-26

    1979

  16. [24]

    Efron, B., 1982, CBMS-NSF Regional Conference Series in Applied Mathematics

    1982

  17. [25]

    2014, A&A, 570, A128

    Ertel, S., Absil, O., Defrère, D., et al. 2014, A&A, 570, A128

    2014

  18. [26]

    2016, A&A, 595, A44

    Ertel, S., Defrère, D., Absil, O., et al. 2016, A&A, 595, A44

    2016

  19. [27]

    Faramaz, V., Ertel, S., Booth, M., Cuadra, J., & Simmonds, C., 2017, MNRAS, 465, 2352

    2017

  20. [28]

    Neural Netw., 1, 179

    Gallant, S., 1990, EEE Trans. Neural Netw., 1, 179

    1990

  21. [29]

    P., Mather, J

    Gardner, J. P., Mather, J. C., Clampin, M., et al., 2006, Space Sci.Rev. 123, 485

    2006

  22. [30]

    G., Rieke, G

    Gáspár, A., Wolff, S. G., Rieke, G. H., et al. 2023, Nature Astronomy, 7, 790-798

    2023

  23. [31]

    15, Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, ed

    Glorot, X., Bordes, A., & Bengio, Y., 2011, PMLR, Vol. 15, Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, ed. G. Gordon, D. Dunson, & M. Dudík (Fort Lauderdale, FL, USA: PMLR), 315–323

    2011

  24. [32]

    Gravity Collaboration (Abuter, R., et al.) 2017, A&A, 602, A94

    2017

  25. [33]

    Hanbury Brown, R., Davis, J., Lake, R. J. W., & Thompson, R. J., 1974, MNRAS, 167, 3

    1974

  26. [34]

    D., 2011, MNRAS, 413, 1515–1523

    Howarth, I. D., 2011, MNRAS, 413, 1515–1523

    2011

  27. [35]

    L., Helton, J

    Iman, R. L., Helton, J. C., & Campbell, J. E., 1981, J. Qual. Technol., 13, 174

    1981

  28. [36]

    R., & Clampin, M., 2005, Nature, 435, 1067

    Kalas, P., Graham, J. R., & Clampin, M., 2005, Nature, 435, 1067

    2005

  29. [37]

    L., Franz, B

    Kelsall, T., Weiland, J. L., Franz, B. A., et al. 1998, ApJ, 508, 44

    1998

  30. [38]

    4006, Interferometry in Optical Astronomy, ed

    Kervella, P., Coudé du Foresto, V., Glindemann, A., & Hofmann, R., 2000, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4006, Interferometry in Optical Astronomy, ed. P. Léna & A. Quirrenbach, 31–42

    2000

  31. [39]

    B., Segransan, D., et al

    Kervella, P., Gitton, P. B., Segransan, D., et al. 2003, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4838, Interferometry for Optical Astronomy II, ed. W. A. Traub, 858–869

    2003

  32. [40]

    Kim, M., Wolf, S., Loehne, T., Kirchschlager, F., & Krivov, A.V., 2018, A&A, 618, A38

    2018

  33. [41]

    K., et al

    Kimura, H., Kunitomo, M., Suzuki, T. K., et al. 2020, Planet. Space Sci., 183, 104581

    2020

  34. [42]

    & Ba, J., 2014, International Conference on Learning Representations

    Kingma, D. & Ba, J., 2014, International Conference on Learning Representations

    2014

  35. [43]

    Kirchschlager F., Wolf S., 2013, A&A, 552, A54

    2013

  36. [44]

    V., Mutschke, H., & Brunngraeber, R., 2017, MNRAS, 467, 1614

    Kirchschlager, F., Wolf, S., Krivov, A. V., Mutschke, H., & Brunngraeber, R., 2017, MNRAS, 467, 1614

    2017

  37. [45]

    V., 2020, MNRAS, 499, L47

    Kirchschlager, F., Ertel, S., Wolf, S., Matter, A., & Krivov, A. V., 2020, MNRAS, 499, L47

    2020

  38. [46]

    Kobayashi, H., Watanabe, S., Kimura, H., & Yamamoto, T., 2009, Icarus, 201, 395

    2009

  39. [47]

    V., Defrère, D., et al

    Kral, Q., Krivov, A. V., Defrère, D., et al. 2017, AR, 13, 69

    2017

  40. [48]

    & Mezard, M., 1987, J

    Krauth, W. & Mezard, M., 1987, J. Phys. A, 20, L745

    1987

  41. [49]

    2019, A&A, 624, A99

    Lacour, S., Dembet, R., Abuter, R., et al. 2019, A&A, 624, A99

    2019

  42. [50]

    C., et al

    Lebreton, J., van Lieshout, R., Augereau, J. C., et al. 2013, A&A, 555, A146

    2013

  43. [51]

    B., Labeye, P., Malbet, F., et al

    Le Bouquin, J. B., Labeye, P., Malbet, F., et al. 2006, A&A, 450, 1259

    2006

  44. [52]

    2014, Messenger, 157, 5

    Lopez, B., Lagarde, S., Jaffe, W., et al. 2014, Messenger, 157, 5

    2014

  45. [53]

    G., et al

    Lopez, B., Lagarde, S., Petrov, R. G., et al. 2022, A&A, 659, A192

    2022

  46. [54]

    A., Matrà, L., Kalas, P., et al

    MacGregor, M. A., Matrà, L., Kalas, P., et al. 2017, ApJ, 842, 8

    2017

  47. [55]

    E., 2012, ApJL, 754, L20

    Mamajek, E. E., 2012, ApJL, 754, L20

    2012

  48. [56]

    P., Cotton, D

    Marshall, J. P., Cotton, D. V., Bott, K., et al. 2016, ApJ, 825, 124

    2016

  49. [57]

    2013, ApJ, 763, 119

    Mennesson, B., Absil, O., Lebreton, J., et al. 2013, ApJ, 763, 119

    2013

  50. [58]

    2014, ApJ, 797, 119

    Mennesson, B., Millan-Gabet, R., Serabyn, E., et al. 2014, ApJ, 797, 119

    2014

  51. [59]

    2011, ApJ, 734, 67

    Millan-Gabet, R., Serabyn, E., Mennesson, B., et al. 2011, ApJ, 734, 67

    2011

  52. [60]

    2024, mfit/lmfit-py: 1.3.2

    Newville, M., Otten, R., Nelson, A., et al. 2024, mfit/lmfit-py: 1.3.2. Zenodo

    2024

  53. [61]

    D., Scott, N

    Nuñez, P. D., Scott, N. J., Mennesson, B., et al. 2017, A&A, 608, A113

    2017

  54. [62]

    Ollmann, K., Wolf, S., Lietzow, M., Stuber, T.A., 2023, A&A, 677, A187

    2023

  55. [63]

    2019, KerasTuner, https://github.com/kerasteam/keras-tuner

    O’Malley, T., Bursztein, E., Long, J., et al. 2019, KerasTuner, https://github.com/kerasteam/keras-tuner

    2019

  56. [64]

    D., Krivov, A

    Pearce, T. D., Krivov, A. V., & Booth, M., 2020, MNRAS, 498, 2798

    2020

  57. [65]

    D., Beust, H., Faramaz, V., et al

    Pearce, T. D., Beust, H., Faramaz, V., et al. 2021, MNRAS, 503, 4

    2021

  58. [66]

    D., Kirchschlager, F., Rouillé, G., et al

    Pearce, T. D., Kirchschlager, F., Rouillé, G., et al. 2022, MNRAS, 517, 1

    2022

  59. [67]

    Reylé, C., Jardine, K., Fouqué, P. et al. 2021, A&A, 650, A201

    2021

  60. [68]

    H., Gáspár, A., & Ballering, N

    Rieke, G. H., Gáspár, A., & Ballering, N. P., 2016, ApJ, 816, 50

    2016

  61. [69]

    Rigley, J. K. & Wyatt, M. C., 2020, MNRAS, 497, 1143

    2020

  62. [70]

    H., Millan-Gabet, R., et al

    Roberge, A., Chen, C. H., Millan-Gabet, R., et al. 2012, PASP, 124, 918

    2012

  63. [71]

    G., 1991, ApJ, 377, 526

    Rouleau, F., & Martin, P. G., 1991, ApJ, 377, 526

    1991

  64. [72]

    B., Argomedo, J., et al

    Schoeller, M., Gitton, P. B., Argomedo, J., et al. 2003, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4838, Interferometry for Optical Astronomy II, ed. W. A. Traub, 870–880

    2003

  65. [73]

    & Glindemann, A., 2003, in ESA Special Publication, Vol

    Schoeller, M. & Glindemann, A., 2003, in ESA Special Publication, Vol. 539, Earths: DARWIN/TPF and the Search for Extrasolar Terrestrial Planets, ed. M. Fridlund, T. Henning, & H. Lacoste, 109–120

    2003

  66. [74]

    Stat., 6, 461-464

    Schwarz, G., 1978, Ann. Stat., 6, 461-464

    1978

  67. [75]

    M., Koresko, C., & Kuchner, M

    Serabyn, E., Mennesson, B., Colavita, M. M., Koresko, C., & Kuchner, M. J., 2012, ApJ, 748, 55

    2012

  68. [76]

    Stamm, J., Czechowski, A., Mann, I., Baumann, C., & Myrvang, M., 2019, A&A, 626, A107

    2019

  69. [77]

    R., Holmes, E

    Stapelfeldt, K. R., Holmes, E. K., Chen, C., et al. 2004, ApJS, 154, 458

    2004

  70. [78]

    C., & Thébault, P., 2019, A&A, 626, A2

    Sezestre, É., Augereau, J. C., & Thébault, P., 2019, A&A, 626, A2

    2019

  71. [79]

    Su, K. Y. L., Rieke, G. H., Malhotra, R., et al. 2013, ApJ, 763, 118

    2013

  72. [80]

    A., Kirchschlager, F., Pearce, T

    Stuber, T. A., Kirchschlager, F., Pearce, T. D., et al. 2023, A&A, 678, A121

    2023

  73. [81]

    Tazzari, M., Beaujean, F., & Testi, L., 2018, MNRAS, 476, 4527

    2018

  74. [82]

    & Doye J., 1997, J

    Wales, D. & Doye J., 1997, J. Phys. Chem., A 101, 5111-5116

    1997

  75. [83]

    Weingartner, J. C. & Draine B. T., 2001, ApJ, 548, 296

    2001

  76. [84]

    2019, A&A, 629, A41

    Woillez, J., Abad, J.A., Abuter, R., et al. 2019, A&A, 629, A41

    2019

  77. [85]

    2024, A&A, 688, A190

    Woillez, J., Petrov, R., Abuter, R., et al. 2024, A&A, 688, A190

    2024

  78. [86]

    & Voshchinnikov, N

    Wolf, S. & Voshchinnikov, N. V., 2004, Comput. Phys. Commun., 162, 113

    2004

  79. [87]

    A., Llop-Sayson, J., et al

    Ygouf, M., Beichman, C. A., Llop-Sayson, J., et al. 2024, Aj, 167, 26

    2024

  80. [88]

    , " * write output.state after.block = add.period write newline

    ENTRY address archiveprefix author booktitle chapter edition editor howpublished institution eprint journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sent...

  81. [89]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

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