pith. sign in

arxiv: 2606.11983 · v1 · pith:77LRIFZQnew · submitted 2026-06-10 · 🌌 astro-ph.IM · astro-ph.EP

Calibration of an Analog-to-Digital Conversion Nonlinearity in JWST/NIRISS

Shashank Dholakia , Shishir Dholakia , Benjamin J. S. Pope , Louis Desdoigts , Shrishmoy Ray , Peter G. Tuthill , Anand Sivaramakrishnan This is my paper

Pith reviewed 2026-06-27 08:21 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.EP
keywords NIRISSADC integral nonlinearityINLflux-dependent systematicramp-fit residualsWASP-39bSOSStransmission spectrum
0
0 comments X

The pith

Periodic ADC nonlinearity in NIRISS data is flux-dependent and correctable to remove 30ppm systematics in spectra.

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

The paper establishes that NIRISS observations show a periodic signal in raw counts with period 1024 ADU whose amplitude grows with pixel flux. This signal is attributed to analog-to-digital converter integral nonlinearity and is modeled by fitting a polynomial-plus-sinusoid function to residuals from ramp fits on uncalibrated frames, yielding an amplitude of 125 ppm. The model is applied to the WASP-39b SOSS dataset, where it eliminates the systematic at the 30 ppm level across both spectral orders and removes a 55 ppm offset between orders. A sympathetic reader cares because the effect is expected in every NIRISS observation and limits the achievable precision for exoplanet atmosphere measurements until corrected in post-processing.

Core claim

The periodic INL is shown to be flux-dependent, increasing in amplitude with higher pixel counts on the detector. We derive a model of this periodic INL by fitting a combination of a polynomial and sinusoid multiplied with the residuals of ramp fits to the uncalibrated data and find an amplitude of 125ppm, up to a 2.5-count shift for a pixel with 20,000ADU. We apply this model to correct the well-studied NIRISS SOSS Program ERS1366 dataset of WASP-39b and reduce the data into a transmission spectrum. We find that our corrected transmission spectrum removes the INL systematic from the uncorrected spectrum at the 30ppm level across both orders, and also corrects a 55ppm offset between Order 1

What carries the argument

Polynomial-plus-sinusoid model of the flux-dependent periodic INL (period 1024 ADU) fitted to ramp-fit residuals on uncalibrated data.

If this is right

  • The correction removes the INL systematic at the 30ppm level across both orders in transmission spectra.
  • A 55ppm offset between Order 1 and Order 2 is eliminated after correction.
  • The effect is present in all NIRISS modes including AMI and SOSS.
  • A larger-scale data-driven calibration is recommended for adoption into NIRISS pipelines.

Where Pith is reading between the lines

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

  • The same correction procedure could be tested on other JWST instruments that use similar detector readout electronics.
  • Incorporating the model into standard calibration pipelines would allow all archival NIRISS data to be re-reduced at higher precision without new observations.
  • Because amplitude scales with flux, the largest corrections apply to bright targets, which may explain why the signal was first noticed in high-signal AMI data.

Load-bearing premise

The observed periodic signal originates from ADC integral nonlinearity and the polynomial-sinusoid fit fully captures the effect without residual biases or mode-specific differences.

What would settle it

Reprocessing a separate set of NIRISS ramp data with the model and verifying that the 1024-ADU periodic component disappears from the residuals at the 30 ppm level would test the claim.

Figures

Figures reproduced from arXiv: 2606.11983 by Anand Sivaramakrishnan, Benjamin J. S. Pope, Louis Desdoigts, Peter G. Tuthill, Shashank Dholakia, Shishir Dholakia, Shrishmoy Ray.

Figure 1
Figure 1. Figure 1: Results from the inference of the integral nonlinearity using Stage 0 NIRISS/SOSS data on WASP-121 b. Residuals to the ramp fitting stage (blue points, top) show a periodic anomaly with a period of roughly 1024 counts. Fitting a nonlinearity model with a polynomial plus a sinusoid proportional to the signal level (green, purple, respectively) to the residuals reduces the sinusoidal anomaly. The corrected d… view at source ↗
Figure 2
Figure 2. Figure 2: Raw NIRISS/SOSS spectrum (top) and difference between the corrected and uncorrected raw spectra (middle) from the first integration for WASP-39 b. Due to our model of the periodic INL as a sinusoid with an amplitude proportional to the signal level, the effects of the correction are stronger in high signal regions of the image, including the spectral trace and structure in the pedestal and bias level. A LS… view at source ↗
Figure 3
Figure 3. Figure 3: Residuals obtained from subtracting the corrected and uncorrected reduced spectra from the NIRISS SOSS data of WASP-39 (corrected-uncorrected). Individual pixels routinely exhibit more than 200 ppm of difference between the corrected and uncorrected data. Different wavelengths have different average pixel counts; the periodic INL thus applies a correction to each wavelength at a different level. Because th… view at source ↗
Figure 4
Figure 4. Figure 4: Transmission spectrum of WASP-39 b with (orange) and without (blue) correction of the inferred periodic integral nonlinearity. With the new correction, the median transit depth is deeper by 25 ppm across Order 1 (circles) and shallower by 30 ppm across Order 2 (triangles). The transit depths exhibit dispersions of 25 ppm and 59 ppm in Orders 1 and 2 respectively, indicating that the periodic INL imparts a … view at source ↗
Figure 5
Figure 5. Figure 5: Standard deviation of transit model residuals as a function of wavelength for the periodic INL-corrected and uncorrected datasets. Bottom panel shows the difference in standard deviations; positive and negative values indicate the correction worsens or improves the scatter respectively. Furthermore, the average ADU count varies signifi￾cantly between and within each spectral order, which can lead to spurio… view at source ↗
read the original abstract

We quantify an unusual flux-dependent systematic which is periodic in raw counts in flight data from the James Webb Space Telescope's Near Infrared Imager and Slitless Spectrograph (JWST/NIRISS), used extensively for exoplanet imaging and spectroscopy. Originally discovered in the aperture masking interferometry (AMI) mode, it also manifests in the Single Object Slitless Spectroscopy (SOSS) mode with the same dominant period of 1024 in raw analog-to-digital units (ADU). The likely cause of the signal is an analog-to-digital converter (ADC) integral nonlinearity (INL) in which case it will apply to all observations taken with the NIRISS instrument. Fortunately, it is straightforward to correct the data in postprocessing. The periodic INL is shown to be flux-dependent, increasing in amplitude with higher pixel counts on the detector. We derive a model of this periodic INL by fitting a combination of a polynomial and sinusoid multiplied with the residuals of ramp fits to the uncalibrated data and find an amplitude of 125ppm, up to a 2.5-count shift for a pixel with 20,000ADU. We apply this model to correct the well-studied NIRISS SOSS Program ERS1366 dataset of WASP-39b and reduce the data into a transmission spectrum. We find that our corrected transmission spectrum removes the INL systematic from the uncorrected spectrum at the 30ppm level across both orders, and also corrects a 55ppm offset between Order 1 and Order 2. We recommend a larger scale data-driven calibration of the periodic INL and the adoption of the outcome into NIRISS data pipelines.

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

3 major / 2 minor

Summary. The manuscript identifies a flux-dependent periodic systematic in JWST/NIRISS data (dominant period 1024 ADU) observed in both AMI and SOSS modes, attributes it to ADC integral nonlinearity, and derives a correction model consisting of a polynomial plus sinusoid fitted to ramp-fit residuals on uncalibrated data (amplitude 125 ppm, up to 2.5 count shift at 20,000 ADU). The model is applied to the ERS1366 WASP-39b SOSS dataset, yielding a transmission spectrum in which the INL systematic is removed at the 30 ppm level across both orders and a 55 ppm offset between Order 1 and Order 2 is corrected. The authors recommend incorporating a larger-scale data-driven calibration into NIRISS pipelines.

Significance. If the correction is shown to be robust and generalizable, the result would be significant for high-precision exoplanet spectroscopy with NIRISS, as it targets a potentially instrument-wide effect at the tens-of-ppm level that affects both imaging and spectroscopic modes. The explicit flux dependence and the practical post-processing approach are strengths. However, the demonstration rests on a single program without reported independent validation, limiting the assessed impact until cross-checks are provided.

major comments (3)
  1. [Abstract] Abstract: the reported 30 ppm removal of the INL systematic and 55 ppm order offset are presented without error bars, uncertainty estimates, or a quantitative assessment of significance, making it impossible to determine whether the improvement exceeds the noise or fit residuals.
  2. [Abstract] Abstract: the polynomial-plus-sinusoid model is obtained by direct fitting to ramp-fit residuals of the uncalibrated data, after which the same functional form is applied to produce the corrected ERS1366 transmission spectrum; this procedure risks circularity because the quoted improvement is measured against a quantity derived from the input data itself.
  3. [Abstract] Abstract: no validation on an independent dataset, no tests for wavelength- or time-dependent residuals, and no exploration of alternative functional forms or sensitivity to the post-hoc choice of polynomial-sinusoid model are described, leaving open the possibility that the reported correction partly reflects overfitting rather than isolation of a true ADC INL term.
minor comments (2)
  1. [Abstract] The abstract states that the periodic signal 'also manifests in the SOSS mode' but provides no quantitative comparison of amplitude or phase between AMI and SOSS, which would strengthen the claim of a common origin.
  2. [Abstract] The recommendation for 'a larger scale data-driven calibration' is appropriate but would benefit from a brief outline of what such a calibration would entail (e.g., number of datasets, fitting strategy) to guide future work.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their detailed and constructive comments on the manuscript. We address each major comment below and indicate the revisions that will be incorporated.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the reported 30 ppm removal of the INL systematic and 55 ppm order offset are presented without error bars, uncertainty estimates, or a quantitative assessment of significance, making it impossible to determine whether the improvement exceeds the noise or fit residuals.

    Authors: We agree that the abstract would benefit from explicit uncertainty estimates and significance assessment. In the revised manuscript we will report the 30 ppm and 55 ppm values together with uncertainties obtained from the spectral extraction covariance and a bootstrap resampling of the binned light curves, and we will state the improvement relative to the per-bin noise floor. revision: yes

  2. Referee: [Abstract] Abstract: the polynomial-plus-sinusoid model is obtained by direct fitting to ramp-fit residuals of the uncalibrated data, after which the same functional form is applied to produce the corrected ERS1366 transmission spectrum; this procedure risks circularity because the quoted improvement is measured against a quantity derived from the input data itself.

    Authors: The model parameters are determined solely from the ramp-fit residuals of the raw, uncalibrated pixel ramps; these residuals are computed before any spectral extraction or binning. The transmission spectrum is subsequently derived from the corrected count-rate images using the standard pipeline steps. The 30 ppm figure therefore quantifies the change in the final science product after an independent correction step. We will revise the abstract and methods to make this workflow and independence explicit. revision: partial

  3. Referee: [Abstract] Abstract: no validation on an independent dataset, no tests for wavelength- or time-dependent residuals, and no exploration of alternative functional forms or sensitivity to the post-hoc choice of polynomial-sinusoid model are described, leaving open the possibility that the reported correction partly reflects overfitting rather than isolation of a true ADC INL term.

    Authors: The periodic signal was first identified in AMI-mode data from a separate program and later confirmed in the SOSS ERS1366 observations; this provides cross-mode consistency for the 1024 ADU period. In the revision we will add explicit checks for wavelength and time dependence of the residuals after correction, and we will report a brief sensitivity test to the choice of polynomial order and sinusoid amplitude. A fully independent SOSS dataset is not analyzed in the present work, which we will note as a limitation while emphasizing that the physical origin (ADC INL) is supported by the shared periodicity across modes. revision: partial

Circularity Check

1 steps flagged

Poly+sinusoid model fitted to ramp residuals of ERS1366 data, then applied to same data's transmission spectrum to claim 30ppm removal

specific steps
  1. fitted input called prediction [Abstract]
    "We derive a model of this periodic INL by fitting a combination of a polynomial and sinusoid multiplied with the residuals of ramp fits to the uncalibrated data and find an amplitude of 125ppm, up to a 2.5-count shift for a pixel with 20,000ADU. We apply this model to correct the well-studied NIRISS SOSS Program ERS1366 dataset of WASP-39b and reduce the data into a transmission spectrum. We find that our corrected transmission spectrum removes the INL systematic from the uncorrected spectrum at the 30ppm level across both orders, and also corrects a 55ppm offset between Order 1 and Order 2."

    The correction parameters are obtained by fitting the poly+sinusoid form to ramp-fit residuals on the uncalibrated ERS1366 data; the claimed 30ppm removal is then measured by applying that same fitted model to derive the corrected spectrum from the identical dataset, so the improvement metric is the direct numerical consequence of the fit rather than an independent test.

full rationale

The paper derives its INL correction by fitting a polynomial-plus-sinusoid model directly to ramp-fit residuals on the uncalibrated ERS1366 dataset, then applies that fitted model to produce a corrected transmission spectrum from the identical observations and reports a 30ppm systematic removal plus 55ppm order offset correction. This reduction is by construction: the reported improvement is the direct effect of subtracting the model fitted to the input residuals, with no independent dataset or hardware validation shown. The assumption that the periodic signal is purely ADC INL is stated but not independently verified. No self-citations, ansatzes, or other load-bearing reductions appear in the provided text.

Axiom & Free-Parameter Ledger

2 free parameters · 1 axioms · 0 invented entities

The ledger is constructed from the abstract description of the fitting process and stated cause; full paper would allow more precise enumeration of parameters.

free parameters (2)
  • INL amplitude = 125 ppm
    Fitted value of 125 ppm reported from the polynomial-sinusoid model on ramp residuals.
  • Dominant period = 1024 ADU
    Observed and used as 1024 ADU in the model.
axioms (1)
  • domain assumption The periodic signal in raw counts is caused by ADC integral nonlinearity
    Abstract states this as the likely cause that applies to all NIRISS observations.

pith-pipeline@v0.9.1-grok · 5876 in / 1618 out tokens · 26305 ms · 2026-06-27T08:21:46.825679+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

40 extracted references · 35 canonical work pages · 4 internal anchors

  1. [1]

    2025a, ApJL, 985, L10, doi: 10.3847/2041-8213/add010

    Ahrer, Radica, Piaulet-Ghorayeb, et al. 2025, ApJL, 985, L10, doi: 10.3847/2041-8213/add010

    work page doi:10.3847/2041-8213/add010 2025

  2. [2]

    The Wide Field Infrared Survey Telescope: 100 Hubbles for the 2020s

    Akeson, Armus, Bachelet, et al. 2019, arXiv e-prints, arXiv:1902.05569, doi: 10.48550/arXiv.1902.05569

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1902.05569 2019

  3. [3]

    2023, PASP, 135, 075001, doi: 10.1088/1538-3873/acd7a3

    Albert, Lafreni` ere, Ren´ e, et al. 2023, PASP, 135, 075001, doi: 10.1088/1538-3873/acd7a3

    work page doi:10.1088/1538-3873/acd7a3 2023

  4. [4]

    2014, Journal of Instrumentation, 9, C03048, doi: 10.1088/1748-0221/9/03/C03048 Astropy Collaboration, Price-Whelan, Lim, et al

    Antilogus, Astier, Doherty, Guyonnet, & Regnault. 2014, Journal of Instrumentation, 9, C03048, doi: 10.1088/1748-0221/9/03/C03048 Astropy Collaboration, Price-Whelan, Lim, et al. 2022, ApJ, 935, 167, doi: 10.3847/1538-4357/ac7c74

    work page doi:10.1088/1748-0221/9/03/c03048 2014

  5. [5]

    2024, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Baines, Espinoza, Filippazzo, Volk, & Carter. 2024, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 13092, Space Telescopes and Instrumentation 2024: Optical, Infrared, and Millimeter Wave, ed. Coyle, Matsuura, & Perrin, 1309212, doi: 10.1117/12.3020330

    work page doi:10.1117/12.3020330 2024

  6. [6]

    2023, Characterization of the visit-to-visit Stability of the GR700XD Spectral Traces for NIRISS/SOSS Observations, Tech

    Baines, Espinoza, Fillipazzo, & Volk. 2023, Characterization of the visit-to-visit Stability of the GR700XD Spectral Traces for NIRISS/SOSS Observations, Tech. Rep. Technical Report JWST-STScI-008448, STScI, doi: 10.48550/arXiv.2311.07769

    work page doi:10.48550/arxiv.2311.07769 2023

  7. [7]

    2011, in Astronomical Society of the Pacific Conference Series, Vol

    Blank, Anglin, Beletic, et al. 2011, in Astronomical Society of the Pacific Conference Series, Vol. 437, Solar Polarization 6, ed. Kuhn, Harrington, Lin, Berdyugina,

    2011

  8. [8]

    2021, arXiv preprint arXiv:2105.15183

    Blondel, Berthet, Cuturi, et al. 2021, arXiv preprint arXiv:2105.15183

    arXiv 2021

  9. [9]

    2018, JAX: composable transformations of Python+NumPy programs, 0.3.13 http://github.com/google/jax

    Bradbury, Frostig, Hawkins, et al. 2018, JAX: composable transformations of Python+NumPy programs, 0.3.13 http://github.com/google/jax

    2018

  10. [10]

    J., et al

    Cadieux, Doyon, MacDonald, et al. 2024, ApJL, 970, L2, doi: 10.3847/2041-8213/ad5afa

    work page doi:10.3847/2041-8213/ad5afa 2024

  11. [11]

    L., May, E

    Carter, May, Espinoza, et al. 2024, Nature Astronomy, 8, 1008, doi: 10.1038/s41550-024-02292-x

    work page doi:10.1038/s41550-024-02292-x 2024

  12. [12]

    2014, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Chen, Loose, Ricardo, et al. 2014, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 9154, High Energy, Optical, and Infrared Detectors for Astronomy VI, ed. Holland & Beletic, 915426, doi: 10.1117/12.2056481

    work page doi:10.1117/12.2056481 2014

  13. [13]

    2024, arXiv e-prints, arXiv:2406.08704, doi: 10.48550/arXiv.2406.08704

    Desdoigts, Pope, Gully-Santiago, & Tuthill. 2024, arXiv e-prints, arXiv:2406.08704, doi: 10.48550/arXiv.2406.08704

    work page doi:10.48550/arxiv.2406.08704 2024

  14. [14]

    AMIGO: a Data-Driven Calibration of the JWST Interferometer

    Desdoigts, Pope, Charles, et al. 2025, arXiv e-prints, arXiv:2510.09806, doi: 10.48550/arXiv.2510.09806

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2510.09806 2025

  15. [15]

    J., Hutchings, J

    Doyon, Willott, Hutchings, et al. 2023, PASP, 135, 098001, doi: 10.1088/1538-3873/acd41b

    work page doi:10.1088/1538-3873/acd41b 2023

  16. [16]

    2011, A&A, 531, A40, doi: 10.1051/0004-6361/201116671

    Faedi, Barros, Anderson, et al. 2011, A&A, 531, A40, doi: 10.1051/0004-6361/201116671

    work page doi:10.1051/0004-6361/201116671 2011

  17. [17]

    D., Radica, M., Welbanks, L., et al

    Feinstein, Radica, Welbanks, et al. 2023, Nature, 614, 670, doi: 10.1038/s41586-022-05674-1

    work page doi:10.1038/s41586-022-05674-1 2023

  18. [18]

    2024, MNRAS, 528, 3354, doi: 10.1093/mnras/stad3813

    Fournier-Tondreau, MacDonald, Radica, et al. 2024, MNRAS, 528, 3354, doi: 10.1093/mnras/stad3813

    work page doi:10.1093/mnras/stad3813 2024

  19. [19]

    2025, MNRAS, 539, 422, doi: 10.1093/mnras/staf489

    Fournier-Tondreau, Pan, Morel, et al. 2025, MNRAS, 539, 422, doi: 10.1093/mnras/staf489

    work page doi:10.1093/mnras/staf489 2025

  20. [20]

    2023, PASP, 135, 068001, doi: 10.1088/1538-3873/acd1b5

    Gardner, Mather, Abbott, et al. 2023, PASP, 135, 068001, doi: 10.1088/1538-3873/acd1b5

    work page doi:10.1088/1538-3873/acd1b5 2023

  21. [21]

    2020, Nature, 585, 357

    Harris, Millman, van der Walt, et al. 2020, Nature, 585, 357

    2020

  22. [22]

    The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo

    Hattori, Garcia, Murray, et al. 2024, exoplanet-dev/jaxoplanet: Astronomical time series analysis with JAX, v0.0.2 Zenodo, doi: 10.5281/zenodo.10736936 Hirata, & Choi. 2020, PASP, 132, 014501, doi: 10.1088/1538-3873/ab44f7 Hoffman, & Gelman. 2011, arXiv e-prints, arXiv:1111.4246, doi: 10.48550/arXiv.1111.4246 Holmberg, & Madhusudhan. 2023, MNRAS, 524, 377...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.5281/zenodo.10736936 2024

  23. [23]

    2024, AJ, 168, 220, doi: 10.3847/1538-3881/ad82ee

    Hviding, Momcheva, & Clarke. 2024, AJ, 168, 220, doi: 10.3847/1538-3881/ad82ee

    work page doi:10.3847/1538-3881/ad82ee 2024

  24. [24]

    , keywords =

    Jones, Oliphant, Peterson, & Others. 2001, SciPy: Open source scientific tools for Python, http://www.scipy.org/ Lomb. 1976, Ap&SS, 39, 447, doi: 10.1007/BF00648343

    work page doi:10.1007/bf00648343 2001

  25. [25]

    2005, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol

    Loose, Beletic, Blackwell, et al. 2005, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 5904, Cryogenic Optical Systems and Instruments XI, ed. Heaney & Burriesci, 293–302, doi: 10.1117/12.619638

    work page doi:10.1117/12.619638 2005

  26. [26]

    2023, ApJL, 956, L13, doi: 10.3847/2041-8213/acf577

    Madhusudhan, Sarkar, Constantinou, et al. 2023, ApJL, 956, L13, doi: 10.3847/2041-8213/acf577 12 Mahalanobis. 1936, Proceedings of the National Institute of Sciences (Calcutta), 2, 49

    work page doi:10.3847/2041-8213/acf577 2023

  27. [27]

    E., Stevenson, K

    Moran, Stevenson, Sing, et al. 2023, ApJL, 948, L11, doi: 10.3847/2041-8213/accb9c Parviainen. 2025, arXiv e-prints, arXiv:2509.15913, doi: 10.48550/arXiv.2509.15913

    work page doi:10.3847/2041-8213/accb9c 2023

  28. [28]

    Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro

    Phan, Pradhan, & Jankowiak. 2019, arXiv preprint arXiv:1912.11554 Radica. 2024, The Journal of Open Source Software, 9, 6898, doi: 10.21105/joss.06898

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.21105/joss.06898 2019

  29. [29]

    2023, Monthly Notices of the Royal Astronomical Society, 524, 835–856, doi: 10.1093/mnras/stad1762

    Radica, Welbanks, Espinoza, et al. 2023, MNRAS, 524, 835, doi: 10.1093/mnras/stad1762

    work page doi:10.1093/mnras/stad1762 2023

  30. [30]

    2025, ApJL, 983, L25, doi: 10.3847/2041-8213/adaeb7

    Ray, Sallum, Hinkley, et al. 2025, ApJL, 983, L25, doi: 10.3847/2041-8213/adaeb7

    work page doi:10.3847/2041-8213/adaeb7 2025

  31. [31]

    2023, PASP, 135, 048001, doi: 10.1088/1538-3873/acb293

    Rigby, Perrin, McElwain, et al. 2023, PASP, 135, 048001, doi: 10.1088/1538-3873/acb293

    work page doi:10.1088/1538-3873/acb293 2023

  32. [32]

    2026, arXiv e-prints, arXiv:2601.02556, doi: 10.48550/arXiv.2601.02556

    Roy, Feldman, Klupar, et al. 2026, arXiv e-prints, arXiv:2601.02556, doi: 10.48550/arXiv.2601.02556

    work page doi:10.48550/arxiv.2601.02556 2026

  33. [33]

    2024, ApJL, 963, L2, doi: 10.3847/2041-8213/ad21fb Scargle

    Sallum, Ray, Kammerer, et al. 2024, ApJL, 963, L2, doi: 10.3847/2041-8213/ad21fb Scargle. 1982, ApJ, 263, 835, doi: 10.1086/160554

    work page doi:10.3847/2041-8213/ad21fb 2024

  34. [34]

    P., MacDonald, R

    Schmidt, MacDonald, Tsai, et al. 2025, AJ, 170, 298, doi: 10.3847/1538-3881/ae019a

    work page doi:10.3847/1538-3881/ae019a 2025

  35. [35]

    2023, PASP, 135, 015003, doi: 10.1088/1538-3873/acaebd

    Sivaramakrishnan, Tuthill, Lloyd, et al. 2023, PASP, 135, 015003, doi: 10.1088/1538-3873/acaebd

    work page doi:10.1088/1538-3873/acaebd 2023

  36. [36]

    2025, arXiv e-prints, arXiv:2509.09760, doi: 10.48550/arXiv.2509.09760

    Splinter, Coulombe, Frazier, et al. 2025, arXiv e-prints, arXiv:2509.09760, doi: 10.48550/arXiv.2509.09760

    work page doi:10.48550/arxiv.2509.09760 2025

  37. [37]

    2018, in American Astronomical Society Meeting Abstracts, Vol

    Stevenson, Fowler, Lewis, et al. 2018, in American Astronomical Society Meeting Abstracts, Vol. 231, American Astronomical Society Meeting Abstracts #231, 148.14

    2018

  38. [38]

    D., et al

    Taylor, Radica, Chatterjee, et al. 2025, MNRAS, 540, 3677, doi: 10.1093/mnras/staf894 VanderPlas. 2018, ApJS, 236, 16, doi: 10.3847/1538-4365/aab766

    work page doi:10.1093/mnras/staf894 2025

  39. [39]

    2026, arXiv e-prints, arXiv:2602.22327, doi: 10.48550/arXiv.2602.22327

    Wallack, Gao, Greklek-McKeon, et al. 2026, arXiv e-prints, arXiv:2602.22327, doi: 10.48550/arXiv.2602.22327

    work page doi:10.48550/arxiv.2602.22327 2026

  40. [40]

    J., Doyon, R., Albert, L., et al

    Willott, Doyon, Albert, et al. 2022, PASP, 134, 025002, doi: 10.1088/1538-3873/ac5158

    work page doi:10.1088/1538-3873/ac5158 2022