Pith. sign in

REVIEW 3 major objections 5 minor 7 cited by

Two JWST epochs reveal 465 variable galactic nuclei in one field, while Little Red Dots stay quiet.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · deepseek-v4-flash

2026-08-04 15:20 UTC pith:6D3HCW2Y

load-bearing objection Solid NEXUS variability catalog and slightly tighter LRD upper limits, but the self-calibrated noise curve deserves scrutiny before the 465 and 3-10% numbers are taken at face value. the 3 major comments →

arxiv 2509.19585 v2 pith:6D3HCW2Y submitted 2025-09-23 astro-ph.GA

NEXUS: A Search for Nuclear Variability with the First Two JWST NIRCam Epochs

Zachary Stone , Yue Shen , Ming-Yang Zhuang , Lei Hu , Justin Pierel , Junyao Li , Adam J. Burgasser , Jenny E. Greene
show 5 more authors
Zhiwei Pan Alice E. Shapley Fengwu Sun Padmavathi Venkatraman Feige Wang
This is my paper
classification astro-ph.GA
keywords nuclear variabilityAGNLittle Red Dotsdifference imagingJWST NIRCamtidal disruption eventsphotometric redshiftsupermassive black holes
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper establishes that difference imaging of two JWST NIRCam epochs separated by nine months can detect nuclear flux changes of 1-2% for bright sources and about 0.1-0.2 magnitudes at magnitude 28. Applying this sensitivity to roughly 25,000 sources, it identifies 465 high-confidence variable sources, almost all extragalactic and likely dominated by active galactic nuclei. It also finds that none of ten spectroscopically confirmed Little Red Dots vary detectably, placing 3σ F444W upper limits of about 3-10% with a median near 5%. This implies weak rest-frame optical variability in Little Red Dots, a direct constraint on accretion models for these enigmatic objects.

Core claim

Using the first two NEXUS epochs (September 2024 and June 2025), the authors match the point-spread function between the two epochs with SFFT and direct subtraction, and adopt the smaller of the two measured magnitude changes as the fiducial value per source. From the 3σ-clipped scatter of this Δm against source magnitude they construct a noise floor, debias the measurements, and select 465 sources whose flux change exceeds 3σ of that floor and survives visual inspection. These variable sources follow the same photometric redshift distribution as the parent sample, indicating they are mostly distant AGNs. Separately, the ten spectroscopically confirmed broad-line Little Red Dots in the field

What carries the argument

The central mechanism is difference imaging with two independent subtractions: SFFT, which models spatial PSF variations and differential background in the Fourier domain, and direct frame subtraction. The 'BEST' measurement combines them by taking the smaller Δm per source. A smooth σ_Δm(m) curve, built from the 3σ-clipped scatter of the same Δm data under the assumption that most sources are non-variable, serves as the noise floor for defining >3σ variability and for setting upper limits on Little Red Dot variability.

Load-bearing premise

The noise curve that sets the 3σ threshold and the Little Red Dot upper limits is measured from the scatter of the same Δm data under the assumption that most sources in each magnitude bin are non-variable; if correlated image-subtraction systematics or a large population of real variables inflate that scatter, both the 465-source catalog and the LRD limits shift.

What would settle it

Compute the Δm scatter on blank sky positions with the same difference-imaging pipeline; if blank-sky noise alone reproduces the σ_Δm(m) curve, the variable-source threshold is noise-limited, while if source positions show excess scatter the curve is contaminated. Alternatively, a third epoch at ~2-month cadence should confirm a large fraction of the 465 candidates as same-sense repeaters; a low repetition fraction would indicate systematics dominated the selection.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • The 465-source catalog provides immediate targets for spectroscopy; the majority should confirm as broad-line AGNs, with a minority possibly being tidal disruption events or supernovae.
  • The 3-10% (median ~5%) F444W non-variability of the ten Little Red Dots constrains models in which the rest-frame optical continuum arises from an extended photosphere or scattered light, both of which smooth variability.
  • Difference imaging cuts the scatter in measured flux changes by more than 30% compared with single-epoch aperture photometry, making it the preferred method for variability searches in future JWST monitoring programs.
  • As NEXUS continues with a 2-month cadence through 2028, the bright subset (F444W < 26) of these variables will become high-priority targets for NIRSpec/MSA spectroscopy to establish their nature.
  • The 465 variables include extreme cases with |Δm| > 1 and off-nucleus flux changes, which are promising candidates for rare nuclear transients and supernovae that can be confirmed with continued monitoring.

Where Pith is reading between the lines

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

  • If the Little Red Dot non-variability persists across the full 18-epoch Deep cadence, it would argue that their rest-frame optical continuum is produced by a spatially extended or self-regulated region rather than a compact, unobscured accretion disk.
  • The same difference-imaging pipeline could be applied to other multi-epoch JWST Treasury fields, and the >30% sensitivity gain over epoch photometry is a directly transferable result for building a uniform census of nuclear variability.
  • With only two epochs, the brightening-versus-dimming asymmetry at faint magnitudes could be used as a statistical flag for transient candidates before spectra arrive; a third epoch will reveal whether the 465 sources repeat in the same sense or represent stochastic AGN flickering.
  • The LRD upper limits, being set on a 9-month observed baseline, correspond to rest-frame time scales of roughly 1-2 months at z~3-7; extending the baseline to several years will probe whether low-level variability appears on longer rest-frame time scales.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. This paper presents a two-epoch JWST/NIRCam variability search over ~25,000 NEXUS sources in F200W and F444W, using difference imaging via direct frame subtraction and the SFFT algorithm. The authors adopt the smaller absolute magnitude change from the two subtraction methods as the fiducial 'BEST' measurement, define variability by |Δm_debiased| > 3σ_Δm, where σ_Δm is a magnitude-dependent noise curve fitted to the 3σ-clipped scatter of the same Δm_BEST data, and after visual inspection produce a catalog of 465 high-confidence variable sources. They also analyze ten spectroscopically confirmed broad-line Little Red Dots, report no significant variability, and derive 3σ F444W upper limits of roughly 3–10% (median ~5%). The paper argues that these limits imply weak rest-frame optical continuum variability in LRDs.

Significance. If the statistical calibration is sound, this is a valuable contribution: it demonstrates the power of NIRCam difference imaging for detecting low-level nuclear variability, provides a public catalog of variable sources at high redshift, and places important constraints on LRD accretion models. The detailed description of the SFFT subtraction pipeline and the careful comparison of methods are strengths. However, the central claims—the 465-source catalog and the quantitative LRD upper limits—rely on a self-calibrated noise curve whose statistical meaning is not well established, so the current version overstates the robustness of the reported thresholds.

major comments (3)
  1. [Section 4, Fig. 7] The variability threshold is calibrated from the same Δm_BEST distribution used for selection. Δm_BEST is the smaller |Δm| of SFFT and direct subtraction; for non-variable sources this is the minimum of two positive-valued noise realizations, which has a narrower, non-Gaussian distribution than the underlying per-method noise. Hence 3σ_Δm does not correspond to a 3σ detection in the original measurement. The initial candidate fraction (~3%) is an order of magnitude higher than the ~0.3% expected for 3σ Gaussian noise, and 36% of candidates are later rejected by visual inspection—both consistent with an underestimated noise floor. The same σ_Δm curve feeds the LRD upper limits in Section 5.2. Please provide an independent noise calibration (e.g., from the scatter of repeated noise realizations, the difference between the two subtraction methods on known non-variables, or injected sources)
  2. [Section 3.4, Eq. (1)] The 'BEST' metric, min(|Δm|) of the two methods, introduces a systematic bias in the measured amplitude for real variables (pulling them toward zero) and is not a standard self-consistent statistic. Its distribution is method-dependent and non-Gaussian, so thresholds based on its clipped standard deviation are not transferable to statements about individual-source significance. This affects both the variable catalog and the upper limits. The authors should either combine the two measurements with a statistically motivated estimator (e.g., inverse-variance weighting after due error calibration) or apply separate thresholds for each method, and should characterize the bias using simulations of injected variability.
  3. [Section 5.2, Fig. 10] The LRD F444W upper limits are directly read off the self-calibrated σ_Δm curve. Given the above calibration concerns, these are not true 3σ upper limits; a factor of 1.5–2 underestimate in the noise would shift the median limit from ~5% toward ~8–10%, weakening the quantitative claim. Also, the text says 'ten LRDs' but Fig. 10 marks two objects without F444W measurements; the authors should explicitly state that only 8 LRDs contribute to the F444W median and clarify the impact on the quoted range.
minor comments (5)
  1. [Section 3.4] The text says 'we adopt the smaller Δm value' but the analysis and figures clearly use the smaller |Δm|. Please use the unambiguous notation throughout.
  2. [Section 4] The threshold for switching from a straight-line fit to a fourth-order polynomial in the σ_Δm(m) curve is not specified. State the magnitude break and its rationale.
  3. [Figure 10] The caption should explain why two LRD sources have no F444W measurements (e.g., outside the F444W coverage or other reasons).
  4. [Section 5.2] When comparing with Kokubo & Harikane (2024) and Zhang et al. (2025), include the quoted limits from those works to make the stated consistency and 'somewhat tighter' claim quantitative.
  5. [Table 1] The table format is dense and the repetition of columns for four methods/apertures may hamper usability. Consider providing a separate machine-readable table with a simplified schema, and point to its column descriptions in the README.

Circularity Check

0 steps flagged

No significant circularity: variability thresholds are empirical noise calibrations, not derived predictions.

full rationale

The paper's central claims are empirical measurements: a catalog of 465 variable sources and 3σ upper limits on LRD variability. The σ_Δm(m) noise curve used for thresholding is fit to the same Δm_BEST data under the explicit assumption that most sources are non-variable (Section 4). This is a standard self-calibration of the noise model, not a circular derivation. The variable selection is defined as |Δm_debiased| > 3σ_Δm, so the curve acts as a significance threshold; the selected sources are outliers relative to the fitted scatter, and 36% are rejected by independent visual inspection. The LRD upper limits are 3σ_Δm values at the LRD magnitudes, i.e., noise-based upper limits, not predictions derived from the fitted parameters. No equation is defined in terms of the result it is supposed to produce, and no fitted constant is renamed as a prediction. Self-citations to NEXUS EDR/overview papers and the LRD spectroscopy paper (Zhuang et al. 2024, 2025; Shen et al. 2024) are data provenance and sample definition, not load-bearing circular arguments. Any concern that the clipped scatter underestimates correlated systematics is a correctness/robustness issue, not a circularity issue. Therefore the derivation chain is self-contained and non-circular.

Axiom & Free-Parameter Ledger

3 free parameters · 4 axioms · 0 invented entities

The paper introduces no new physical entities. Its free parameters are empirical noise-curve fits and image-processing thresholds, which are internal calibrations. The key external assumptions are that the majority of sources are non-variable and that visual inspection reliably separates true variables from subtraction artifacts.

free parameters (3)
  • sigma_Delta-m(m) noise-curve coefficients = not quoted in draft
    A straight-line fit at bright magnitudes and a fourth-order polynomial at faint magnitudes are fitted to the 3-sigma-clipped binned Delta-m scatter (Section 4). This curve defines the 3-sigma variability threshold and the LRD upper limits.
  • SFFT masking thresholds = FLUX_APER = 30, 45, 150 microJy; CLASS_STAR = 0.98, 0.9, 0.7
    Hand-chosen source-masking thresholds in Section 3.2, described as 'have been shown to work best', control which pixels SFFT fits and therefore affect residual noise and the variable-source sample.
  • SFFT algorithm hyperparameters = kernel half-widths 11 px F200W and 5 px F444W; 3x3 and 6x6 B-spline knot grids; Tikhonov lambda = 3e-5; first-order phot
    These model choices affect PSF matching quality and background modeling, and thus the Delta-m scatter. They are chosen from testing, not fitted to the variability signal itself.
axioms (4)
  • domain assumption The majority of sources in each magnitude bin are non-variable, so the 3-sigma-clipped scatter of Delta-m equals the measurement noise floor.
    Stated in Section 4: 'assuming the majority of sources are non-variable'. The variability threshold and LRD upper limits inherit this assumption.
  • domain assumption Residual PSF-matching and background artifacts are either contained within the 0.2 arcsecond aperture and do not bias aperture fluxes, or are correctly removed by the authors' visual inspection.
    Section 4 prunes 36 percent of automated candidates by eye; the final sample of 465 depends on this screening being accurate and repeatable.
  • domain assumption Relative astrometry with RMS of 8-9 mas between epochs and filters is sufficient for 0.2 arcsecond apertures to capture the nuclear flux.
    Section 2.2; if alignment is worse for faint or extended sources, difference-flux apertures can lose or mix flux and inflate apparent variability.
  • domain assumption F444W traces the rest-frame optical continuum of LRDs at z~3-7, so non-variability in F444W constrains optical continuum variability.
    Section 5.2; this is a standard SED assumption for LRDs, not a measurement made in this paper.

pith-pipeline@v1.3.0-alltime-deepseek · 17263 in / 10049 out tokens · 77034 ms · 2026-08-04T15:20:44.994734+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of NEXUS: A Search for Nuclear Variability with the First Two JWST NIRCam Epochs." pith.science (2026). https://pith.science/paper/6D3HCW2Y

@misc{pith2026250919585,
  author       = {Pith},
  title        = {Pith review of: NEXUS: A Search for Nuclear Variability with the First Two JWST NIRCam Epochs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6D3HCW2Y}},
  note         = {Machine review of arXiv:2509.19585}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

The multi-cycle JWST Treasury program NEXUS will obtain cadenced imaging and spectroscopic observations around the North Ecliptic Pole during 2024-2028. Here we report a systematic search for nuclear variability among $\sim 25\,$k sources covered by NIRCam (F200W+F444W) imaging using the first two NEXUS epochs separated by 9 months in the observed frame. Difference imaging techniques reach $1\sigma$ variability sensitivity of 0.18~mag (F200W) and 0.15~mag (F444W) at 28th magnitude (within 0".2 diameter aperture), improved to $0.01$~mag and $0.02$~mag at $<25$th magnitude, demonstrating the superb performance of NIRCam photometry. The difference imaging results represent significant improvement over aperture photometry on individual epochs (by $>30\%$). We identify 465 high-confidence variable sources among the parent sample, with 2-epoch flux difference at $>3\sigma$ from the fiducial variability sensitivity. Essentially all these variable sources are of extragalactic origin based on preliminary photometric classifications, and follow a similar photometric redshift distribution as the parent sample up to $z_{\rm phot}>10$. While the majority of these variability candidates are likely normal unobscured AGNs, some of them may be rare nuclear stellar transients and tidal disruption events that await confirmation with spectroscopy and continued photometric monitoring. We also constrain the photometric variability of ten spectroscopically confirmed broad-line Little Red Dots (LRDs) at $3\lesssim z \lesssim 7$, and find none of them show detectable variability in either band. We derive stringent $3\sigma$ upper limits on the F444W variability of $\sim 3-10\%$ for these LRDs, with a median value of $\sim 5\%$. These constraints imply weak variability in the rest-frame optical continuum of LRDs.

Figures

Figures reproduced from arXiv: 2509.19585 by Adam J. Burgasser, Alice E. Shapley, Feige Wang, Fengwu Sun, Jenny E. Greene, Junyao Li, Justin Pierel, Lei Hu, Ming-Yang Zhuang, Padmavathi Venkatraman, Yue Shen, Zachary Stone, Zhiwei Pan.

Figure 1
Figure 1. Figure 1: NIRCam imaging overlap between the first (par￾tial) Wide epoch (Wide-1.1; background image) and the first Deep epoch (Deep-1; colored mosaics for different pointings). stage 2 image calibration, and stage 3 mosaic construc￾tion. Prior to stage 3, we remove a pedestal background, “1/f” noise, and large-scale “wisp” emission, and mask other artifacts caused by scattered light1 . We aggres￾sively mask the sou… view at source ↗
Figure 2
Figure 2. Figure 2: The procedure used to generate our difference images, displaying an example 900×900 pixel cutout in the F444W band. The sky-subtracted cutouts are shown in the two upper left panels, and the cross-convolved cutouts are shown directly beneath them. The PSFs used for each of the input images is shown in an inset in the two upper left panels, along with their position angles. In the top right, three panels sh… view at source ↗
Figure 3
Figure 3. Figure 3: A few example source cutouts in both NEXUS epochs, as well as the difference images from both subtraction methods to compare their performance. For each source, the top (bottom) row represents images in the F200W (F444W) band. The image shown for each column is displayed at the top of each column. The color scale used for the difference images ranges from ±5σ of the background pixels, with blue representin… view at source ↗
Figure 4
Figure 4. Figure 4: A comparison of the magnitude changes ∆m of sources identified within the NEXUS field between both subtraction methods, with mDIFF,BEST < 35 using a 0. ′′2 aperture. The row of each panel corresponds to a different filter, and each column of a panel corresponds to a different aperture, labeled at the top of each column. Solid lines represent an equal ∆m between the two methods. Black circles correspond to … view at source ↗
Figure 5
Figure 5. Figure 5: A comparison of the scatter (standard deviation) σ∆m of the ∆m distribution as a function of the average source magnitude across two epochs ⟨m⟩ between different methods, using an aperture of 0. ′′2. Each panel shows a dif￾ferent band, displayed in its upper left corner. Each color line represents a different method, as indicated in the up￾per right legend. The “BEST” results are the smaller ∆m value from … view at source ↗
Figure 6
Figure 6. Figure 6: Histograms comparing the distribution of ∆m for different methods, within different ⟨m⟩ bins. Each row represents a unique band, while each column represents a ⟨m⟩ bin. Each color represents a different method, identified using the legend in the upper right corner. Note that the spikes at ∆m ≈ 0 from difference imaging methods are dominated by well-subtracted sources with difference mag > 35. These sources… view at source ↗
Figure 7
Figure 7. Figure 7: The distribution of ∆mBEST for all considered bands (rows), using a 0. ′′2 aperture. The 3σ-clipped, binned (∆m mean, σ∆m, ∆merr median) values are shown using (black, green, magenta) circles. Dashed yellow lines represent the (1,2,3)σ deviation from the mean. The solid blue line represents the 1σ deviation from the mean using the epoch photometry ∆m. The red points are sources with |∆mdebiased| > 3σ∆m. Li… view at source ↗
Figure 8
Figure 8. Figure 8: Distribution of the sample of high-confidence nuclear variables in the photometric redshift - mF444W plane, color-coded by the maximum variability in the two bands. These nuclear variables follow a similar photo-z distribution as the underlying sample (gray points). ables in the photometric redshift versus F444W magni￾tude plane is shown in [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: A few example source cutouts (2′′×1 ′′) with significant variability detected by our approach. For each source, each column corresponds to the reference, science or “BEST” difference image. The difference image uses a linear color scale, where blue is negative, red is positive, and the limits of the color map are ±5σ of the background pixels. A 0. ′′2 diameter aperture is drawn around the source in each im… view at source ↗
Figure 10
Figure 10. Figure 10: 3σ upper limit of variability for the 10 spectroscopically-confirmed LRDs in Zhuang et al. (2025) based on our variability analysis. Two objects do not have F444W variability measurements and are marked as open circles at the bottom. The dashed lines mark the median value of the 3σ upper limits. the sensitivity of variability detection using difference imaging over epoch photometry. Our best performance r… view at source ↗

discussion (0)

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

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. NEXUS: Spectral Variability of Little Red Dots and Blue Active Galactic Nuclei at $2 \lesssim z \lesssim 6$

    astro-ph.GA 2026-08 conditional novelty 6.0

    Little Red Dots at z=2-6 show less than 4% intrinsic H-alpha variability on 1-3 month rest-frame timescales, a flat white-noise pattern unlike normal AGNs, implying different broad-line production.

  2. ATLAS. II. Extremely High Incidence of Balmer Line Absorption with Predominant Blueshifts in LRDs: Statistical Insights through Comparison with Type 1 AGNs

    astro-ph.GA 2026-07 conditional novelty 6.0

    Balmer-line absorption occurs in ~35% (14/40) of JWST little-red-dot AGNs, roughly 850x the rate in SDSS type-1 AGNs, with mostly slow blueshifted absorber velocities.

  3. Little Red Dots as Intermediate Mass, Super-Eddington Engines: Insights from Type IIn Supernovae and The 1837-1856 Great Eruption of $\eta$ Carinae

    astro-ph.GA 2026-06 unverdicted novelty 6.0

    LRDs are reinterpreted as intermediate-mass super-Eddington systems with wind-driven pseudo-photospheres that explain their spectra and imply engine masses below 10^5 solar masses rather than overmassive black holes.

  4. A Scaling Relation of LRDs between Broad H$\alpha$ and Bolometric Luminosities: Enhanced Broad H$\alpha$ Emission Relative to Low-$z$ Type 1 AGN

    astro-ph.GA 2026-06 unverdicted novelty 6.0

    LRDs at z~3-7 exhibit an L_Hα,broad-L_bol scaling relation enhanced by a factor of ~40 compared to low-z Type 1 AGN, explained via Cloudy modeling with near-unity covering factor and high column density.

  5. NEXUS: Abundance, Environments, and Spectral Diversity of Little Red Dots from the NIRSpec MSA Sample

    astro-ph.GA 2026-06 unverdicted novelty 5.0

    A sample of 36 spectroscopically confirmed LRDs shows broad-line detections in >90%, spectral variety including Balmer breaks and blackbody fits, H-alpha to 5100A continuum correlation, no redshift evolution, declinin...

  6. Mass and Spin Growth of Very Massive Stars in Star Clusters Potentially Associated with Little Red Dots

    astro-ph.HE 2026-06 unverdicted novelty 4.0

    Simulations show VMS in star clusters reach 10^3-10^4 solar masses with dimensionless spins >10 under bloated accretion conditions, potentially forming spinning IMBHs that produce GW bursts like GW190521.

  7. Unveil the nature of JWST-AGN and Little Red Dots with SKAO continuum surveys

    astro-ph.GA 2026-06 unverdicted novelty 3.0

    SKAO continuum surveys will detect radio emission from JWST AGN and LRDs and distinguish between Compton-thick absorption, intrinsically weak accretion, and dense gas cocoon scenarios.

Reference graph

Works this paper leans on

33 extracted references · 2 canonical work pages · cited by 7 Pith papers · 1 internal anchor

  1. [1]

    2015, The Astrophysical Journal Supplement Series, 220, 1, doi: 10.1088/0067-0049/220/1/1 Astropy Collaboration, Robitaille, T

    Akhlaghi, M., & Ichikawa, T. 2015, The Astrophysical Journal Supplement Series, 220, 1, doi: 10.1088/0067-0049/220/1/1 Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f

    doi:10.1088/0067-0049/220/1/1 2015

  2. [2]

    1996, Astronomy and Astrophysics Supplement Series, 117, 393, doi: 10.1051/aas:1996164

    Bertin, E., & Arnouts, S. 1996, Astronomy and Astrophysics Supplement Series, 117, 393, doi: 10.1051/aas:1996164

    doi:10.1051/aas:1996164 1996

  3. [3]

    2002, in Astronomical Society of the Pacific Conference Series, Vol

    Bertin, E., Mellier, Y., Radovich, M., et al. 2002, in Astronomical Society of the Pacific Conference Series, Vol. 281, Astronomical Data Analysis Software and Systems XI, ed. D. A. Bohlender, D. Durand, & T. H. Handley, 228

    2002

  4. [4]

    2022, astropy/photutils: 1.5.0, 1.5.0, Zenodo, doi: 10.5281/zenodo.6825092

    Bradley, L., Sip˝ ocz, B., Robitaille, T., et al. 2022, astropy/photutils: 1.5.0, 1.5.0, Zenodo, doi: 10.5281/zenodo.6825092

    doi:10.5281/zenodo.6825092 2022

  5. [5]

    M., Kartaltepe, J

    Casey, C. M., Kartaltepe, J. S., Drakos, N. E., et al. 2023, ApJ, 954, 31, doi: 10.3847/1538-4357/acc2bc

    doi:10.3847/1538-4357/acc2bc 2023

  6. [6]

    DeCoursey, C., Egami, E., Pierel, J. D. R., et al. 2025a, The Astrophysical Journal, 979, 250, doi: 10.3847/1538-4357/ad8fab

    doi:10.3847/1538-4357/ad8fab

  7. [7]

    2025b, The Astrophysical Journal, 990, 31, doi: 10.3847/1538-4357/ade78c

    DeCoursey, C., Egami, E., Sun, F., et al. 2025b, The Astrophysical Journal, 990, 31, doi: 10.3847/1538-4357/ade78c

    doi:10.3847/1538-4357/ade78c

  8. [8]

    J., Willott, C., Alberts, S., et al

    Eisenstein, D. J., Willott, C., Alberts, S., et al. 2023, arXiv e-prints, arXiv:2306.02465, doi: 10.48550/arXiv.2306.02465

    Pith/arXiv arXiv doi:10.48550/arxiv.2306.02465 2023

  9. [9]

    E., Labbe, I., Goulding, A

    Greene, J. E., Labbe, I., Goulding, A. D., et al. 2024, ApJ, 964, 39, doi: 10.3847/1538-4357/ad1e5f

    doi:10.3847/1538-4357/ad1e5f 2024

  10. [10]

    N., Maiolino, R., Juodˇ zbalis, I., et al

    Hainline, K. N., Maiolino, R., Juodˇ zbalis, I., et al. 2025, ApJ, 979, 138, doi: 10.3847/1538-4357/ad9920

    doi:10.3847/1538-4357/ad9920 2025

  11. [11]

    R., Millman, K

    Harris, C. R., Millman, K. J., van der Walt, S. J., et al. 2020, Nature, 585, 357, doi: 10.1038/s41586-020-2649-2

    doi:10.1038/s41586-020-2649-2 2020

  12. [12]

    2024, The Astronomical Journal, 167, 231, doi: 10.3847/1538-3881/ad36cb

    Hu, L., & Wang, L. 2024, The Astronomical Journal, 167, 231, doi: 10.3847/1538-3881/ad36cb

    work page doi:10.3847/1538-3881/ad36cb 2024

  13. [13]

    2022, The Astrophysical Journal, 936, 157, doi: 10.3847/1538-4357/ac7394

    Hu, L., Wang, L., Chen, X., & Yang, J. 2022, The Astrophysical Journal, 936, 157, doi: 10.3847/1538-4357/ac7394

    doi:10.3847/1538-4357/ac7394 2022

  14. [14]

    Hunter, J. D. 2007, Computing in Science & Engineering, 9, 90, doi: 10.1109/MCSE.2007.55 16

    doi:10.1109/mcse.2007.55 2007

  15. [15]

    D., Finkelstein, S

    Kocevski, D. D., Finkelstein, S. L., Barro, G., et al. 2024, arXiv e-prints, arXiv:2404.03576, doi: 10.48550/arXiv.2404.03576

    Pith/arXiv arXiv doi:10.48550/arxiv.2404.03576 2024

  16. [16]

    2024, arXiv e-prints, arXiv:2407.04777, doi: 10.48550/arXiv.2407.04777

    Kokubo, M., & Harikane, Y. 2024, arXiv e-prints, arXiv:2407.04777, doi: 10.48550/arXiv.2407.04777

    doi:10.48550/arxiv.2407.04777 2024

  17. [17]

    E., Bezanson, R., et al

    Labbe, I., Greene, J. E., Bezanson, R., et al. 2025, ApJ, 978, 92, doi: 10.3847/1538-4357/ad3551

    doi:10.3847/1538-4357/ad3551 2025

  18. [18]

    E., & Ma, Y

    Liu, H., Jiang, Y.-F., Quataert, E., Greene, J. E., & Ma, Y. 2025, doi: 10.48550/arXiv.2507.07190

    doi:10.48550/arxiv.2507.07190 2025

  19. [19]

    P., Brammer, G., et al

    Matthee, J., Naidu, R. P., Brammer, G., et al. 2024, ApJ, 963, 129, doi: 10.3847/1538-4357/ad2345

    doi:10.3847/1538-4357/ad2345 2024

  20. [20]

    D., Sivaramakrishnan, A., Lajoie, C.-P., et al

    Perrin, M. D., Sivaramakrishnan, A., Lajoie, C.-P., et al. 2014, in Space Telescopes and Instrumentation 2014:

    2014

  21. [21]

    9143, 91433X, doi: 10.1117/12.2056689

    Optical, Infrared, and Millimeter Wave, Vol. 9143, 91433X, doi: 10.1117/12.2056689

    doi:10.1117/12.2056689

  22. [22]

    S., Jiang, Y.-F., et al

    Secunda, A., Somerville, R. S., Jiang, Y.-F., et al. 2025, doi: 10.48550/arXiv.2509.03571

    Pith/arXiv arXiv doi:10.48550/arxiv.2509.03571 2025

  23. [23]

    2024, arXiv e-prints, arXiv:2408.12713, doi: 10.48550/arXiv.2408.12713

    Shen, Y., Zhuang, M.-Y., Li, J., et al. 2024, arXiv e-prints, arXiv:2408.12713, doi: 10.48550/arXiv.2408.12713

    Pith/arXiv arXiv doi:10.48550/arxiv.2408.12713 2024

  24. [24]

    2025, doi: 10.48550/arXiv.2503.15587

    Sun, F., Fudamoto, Y., Lin, X., et al. 2025, doi: 10.48550/arXiv.2503.15587

    Pith/arXiv arXiv doi:10.48550/arxiv.2503.15587 2025

  25. [25]

    J., Barger, A

    Taylor, A. J., Barger, A. J., Cowie, L. L., et al. 2023, ApJS, 266, 24, doi: 10.3847/1538-4365/accd70

    doi:10.3847/1538-4365/accd70 2023

  26. [26]

    E., et al

    Virtanen, P., Gommers, R., Oliphant, T. E., et al. 2020, Nature Methods, 17, 261, doi: 10.1038/s41592-019-0686-2

    doi:10.1038/s41592-019-0686-2 2020

  27. [27]

    A., Cohen, S

    Windhorst, R. A., Cohen, S. H., Jansen, R. A., et al. 2023, AJ, 165, 13, doi: 10.3847/1538-3881/aca163

    doi:10.3847/1538-3881/aca163 2023

  28. [28]

    2023, ApJS, 269, 43, doi: 10.3847/1538-4365/ad0298

    Yan, H., Ma, Z., Sun, B., et al. 2023, ApJS, 269, 43, doi: 10.3847/1538-4365/ad0298

    doi:10.3847/1538-4365/ad0298 2023

  29. [29]

    Zhang, Z., Jiang, L., Liu, W., & Ho, L. C. 2024, arXiv e-prints, arXiv:2411.02729, doi: 10.48550/arXiv.2411.02729

    Pith/arXiv arXiv doi:10.48550/arxiv.2411.02729 2024

  30. [30]

    Zhang, Z., Jiang, L., Liu, W., & Ho, L. C. 2025, The Astrophysical Journal, 985, 119, doi: 10.3847/1538-4357/adcb3e

    doi:10.3847/1538-4357/adcb3e 2025

  31. [31]

    Zhou, S., Sun, M., Zhang, Z., Chen, J., & Ho, L. C. 2025, arXiv e-prints, arXiv:2508.16795, doi: 10.48550/arXiv.2508.16795

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2508.16795 2025

  32. [32]

    2025, doi: 10.48550/arXiv.2505.20393

    Zhuang, M.-Y., Li, J., Shen, Y., et al. 2025, doi: 10.48550/arXiv.2505.20393

    Pith/arXiv arXiv doi:10.48550/arxiv.2505.20393 2025

  33. [33]

    2024, arXiv e-prints, arXiv:2411.06372, doi: 10.48550/arXiv.2411.06372

    Zhuang, M.-Y., Wang, F., Sun, F., et al. 2024, arXiv e-prints, arXiv:2411.06372, doi: 10.48550/arXiv.2411.06372

    Pith/arXiv arXiv doi:10.48550/arxiv.2411.06372 2024