pith. sign in

arxiv: 2509.06929 · v3 · pith:FY2KQKO2new · submitted 2025-09-08 · 🌌 astro-ph.CO · astro-ph.IM

Edges In Coadded Images

Erin Sheldon This is my paper

Pith reviewed 2026-05-21 22:29 UTC · model grok-4.3

classification 🌌 astro-ph.CO astro-ph.IM
keywords weak gravitational lensingcoadded imagespoint spread functionshear measurementimage edgesmetacalibrationmetadetection
0
0 comments X

The pith

Coadd image edges introduce no significant bias to weak lensing shear measurements under typical survey conditions.

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

The paper tests whether discontinuities in the point spread function and noise, created when edges from input images cross coadd regions, bias shear estimates in weak gravitational lensing. Simulations using Metacalibration and Metadetection for calibration and a simple weighted mean coaddition show no significant biases for typical galaxy populations, edge hit rates of a few percent, or coadds built from tens to hundreds of input images. Significant biases that exceed large-survey requirements appear only in extreme cases such as coadds from just two images, frequent edge crossings, PSF size changes over 25 percent, or small galaxies. Even then a statistic that measures relative PSF size variation across each object can flag and remove the affected measurements to recover accurate shear.

Core claim

Discontinuities in the point spread function and image noise at edges in coadded images do not produce significant biases in weak lensing shear measurements for typical survey conditions and galaxy populations. Biases arise only in extreme scenarios such as coadds made from just two images or when edges cross objects frequently, and these can be mitigated by flagging measurements with large relative PSF size variations.

What carries the argument

A simple statistic that quantifies the relative variation in PSF size across each object, used to identify and remove measurements affected by edge discontinuities.

If this is right

  • Typical coadds with tens to hundreds of input images and low edge hit rates yield unbiased shear measurements.
  • Extreme conditions such as two-input coadds or 25 percent edge crossings require additional flagging via the PSF variation statistic.
  • Accurate shear recovery remains possible by removing measurements identified by large relative PSF size changes.
  • Biases exceed requirements only for small galaxies combined with large PSF size variations.

Where Pith is reading between the lines

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

  • Survey pipelines could adopt the PSF variation statistic as a routine quality cut without major changes to coaddition methods.
  • Repeating the test on real survey data could reveal additional unmodeled correlations at edges not present in the simulations.
  • The same flagging approach may help control PSF-related systematics in other galaxy shape measurements beyond lensing.

Load-bearing premise

The simulations accurately capture the statistical properties of real survey PSF discontinuities and noise at image edges.

What would settle it

Finding significant shear biases in real coadded images from a large survey with typical edge hit rates of a few percent would show that the simulations missed important effects.

Figures

Figures reproduced from arXiv: 2509.06929 by Erin Sheldon.

Figure 1
Figure 1. Figure 1: Example PSF ellipticity whiskers from the realistic simulation, showing optics (left panel), atmo [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Distribution of PSF FWHM in the realistic simulation. The FWHM was measured on the full [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Example PSF FWHM in arcsec from the realistic simulation, showing optics (left panel), atmo [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Example postage stamps from the realistic simulations. Upper left: the coadded image. Upper [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Multiplicative bias m as a function of T /TPSF for various simulation settings. In the legend, e2 indicates two epochs, or CCD images, contributing to the coadd, e30 indicates 30 epochs etc. fixedgal indicates the fixed exponential model, for which all galaxies had a fixed size, and the associated “no edges” or 0.02 value indicates the rate of edge hits. wldb+pointings indicates the realistic simulation wi… view at source ↗
Figure 6
Figure 6. Figure 6: Multiplicative bias m as a function of the rate of edge hits in the coadd for various simulation settings, including variation of the PSF FWHM standard deviation σ. See figure 5 for other label definitions. A bias was found for the case of 2 epochs per coadd and σ = 0.2 arcsec. Cutting postage stamps with relatively large PSF size variations due to edges Tfrac=σ(T PSF)/⟨T PSF⟩< 0.01 removes the bias. The w… view at source ↗
Figure 7
Figure 7. Figure 7: Correlation functions ξ+ and ξ− for zero shear simulations. Top row: A simulation with 300 epochs, on average, contributing to the coadd at a given position. Middle row: A simulation with an average of 30 epochs contributing to the coadds. Bottom row: A simulation with an average of 2 epochs contributing to the coadds. No significant correlations were detected, and the upper bound on the measurement is wel… view at source ↗
Figure 8
Figure 8. Figure 8: Distribution of Tfrac=σ(T PSF)/⟨T PSF⟩ for various simulation configurations. For the realistic LSST-like simulations with 460 epochs, similar to year 10, 3-band coadds, all values of Tfrac are less than 0.002, which is negligible. For 30 epochs, higher Tfrac values were found, but only 0.3% of Tfrac exceeded 0.01. The more extreme, targeted simulations with higher edge hit rate and larger PSF variation pr… view at source ↗
read the original abstract

We investigate how discontinuities in the point spread function (PSF) and image noise affect weak gravitational lensing shear measurements. Our focus is on discontinuities that arise in coadded images, particularly when edges from input images cross the coadd region. Using Metacalibration and Metadetection for shear calibration and a simple weighted mean coaddition, we find no significant biases for typical galaxy populations, typical edge hit rates (a few percent), or coadds with tens to hundreds of input images. Biases exceeding requirements for large lensing surveys occur only in extreme conditions: (a) coadds with just two input images; (b) an image edge crosses the object in about 25% of coadds; (c) PSF size variations greater than 25%; (d) relatively small galaxies. Even in these extreme cases, accurate shear recovery is achievable by identifying and removing problematic measurements. We use a simple statistic that quantifies the relative variation in PSF size across each object.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 2 minor

Summary. The paper investigates the impact of discontinuities in the point spread function (PSF) and image noise at the edges of input images within coadded images on weak gravitational lensing shear measurements. Employing simulations of a simple weighted mean coaddition along with Metacalibration and Metadetection for shear calibration, the authors find no significant biases for typical galaxy populations, typical edge hit rates (a few percent), or coadds with tens to hundreds of input images. Biases that exceed requirements for large surveys occur only under extreme conditions such as coadds with only two input images, an image edge crossing the object in about 25% of cases, PSF size variations greater than 25%, or relatively small galaxies. They introduce a simple statistic quantifying the relative variation in PSF size across each object to identify and remove problematic measurements.

Significance. If the simulations faithfully represent real survey conditions, this result is of high significance for weak lensing analyses in surveys like LSST and Euclid. It indicates that coadd edges do not pose a major systematic for typical data, allowing focus on other calibration issues. The controlled simulation approach is a strength, enabling clear isolation of edge effects and identification of when they become problematic. The proposal of a diagnostic statistic adds practical value for data quality assessment.

major comments (2)
  1. [Simulation Setup] The simulations model PSF discontinuities and noise edges in weighted-mean coadds, but there is no quantitative comparison provided to the statistical properties of real survey coadds, such as histograms of PSF size variations or noise power spectra at edges from datasets like DES or HSC. This validation is load-bearing for the central claim that there are 'no significant biases for typical galaxy populations' and 'typical edge hit rates (a few percent)', because if real data have additional unmodeled correlations from dither patterns or interpolation kernels, the conclusion would not hold.
  2. [Results and Discussion] For the extreme cases (e.g., two input images or 25% edge hit rate), the manuscript should include specific numerical values for the measured shear biases and how they compare to the survey requirements, perhaps with error bars from the simulations, to allow assessment of the severity.
minor comments (2)
  1. [Abstract] The abstract refers to 'a simple statistic' without providing its name or a brief definition; including this would enhance readability.
  2. [Figures] Ensure that figures showing bias as a function of edge hit rate or PSF variation have clear labels for the 'typical' regime versus 'extreme'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our manuscript. We appreciate the positive assessment of the work's significance for surveys such as LSST and Euclid. Below we respond point by point to the major comments, indicating where revisions will be made.

read point-by-point responses

  1. Referee: [Simulation Setup] The simulations model PSF discontinuities and noise edges in weighted-mean coadds, but there is no quantitative comparison provided to the statistical properties of real survey coadds, such as histograms of PSF size variations or noise power spectra at edges from datasets like DES or HSC. This validation is load-bearing for the central claim that there are 'no significant biases for typical galaxy populations' and 'typical edge hit rates (a few percent)', because if real data have additional unmodeled correlations from dither patterns or interpolation kernels, the conclusion would not hold.

    Authors: We agree that direct quantitative comparisons to real survey coadds would strengthen the applicability of our results. Our simulations were intentionally constructed as a controlled, minimal model using weighted-mean coaddition to isolate the effects of PSF size jumps and noise discontinuities at edges, without confounding factors from complex pipelines. The input parameters (edge hit rates of a few percent, PSF variations, number of input images) were selected to bracket values commonly reported for DES and HSC coadds in the literature. We will revise the manuscript to include a new subsection in the discussion that compares our simulated distributions of PSF size variations to published histograms from DES and HSC coadd analyses, citing the relevant references. We will also explicitly note the limitation that our model does not include potential additional correlations from dither patterns or interpolation kernels, and discuss why such effects are expected to be subdominant for stacks with tens to hundreds of inputs. This addresses the validation concern while preserving the paper's focus on isolating the edge discontinuity effect. revision: partial

  2. Referee: [Results and Discussion] For the extreme cases (e.g., two input images or 25% edge hit rate), the manuscript should include specific numerical values for the measured shear biases and how they compare to the survey requirements, perhaps with error bars from the simulations, to allow assessment of the severity.

    Authors: We agree that explicit numerical values will improve the clarity and utility of the results for readers. The current manuscript reports that biases exceed survey requirements only in the listed extreme cases but does not tabulate the precise values. In the revised version we will add a table (or expanded figure caption) reporting the measured multiplicative bias m and additive bias c for each extreme configuration, including the 1-sigma uncertainties derived from the simulation ensembles. These will be directly compared to the LSST and Euclid requirements (e.g., |m| < 0.001 and |c| < 0.0001 per component). This addition will allow quantitative assessment of severity without altering the paper's conclusions. revision: yes

Circularity Check

0 steps flagged

No circularity; simulation-driven results independent of inputs

full rationale

The paper reports empirical results from controlled simulations of coadded images using standard Metacalibration and Metadetection pipelines. No equations, fitted parameters, or self-citations are presented that reduce the central claims (no significant bias for typical edge rates) to tautological re-statements of the simulation setup. The work compares simulated shear biases against external survey requirements rather than deriving them from prior author results or definitional loops. This is the common case of a self-contained simulation study.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; no explicit free parameters, axioms, or invented entities are stated. The central claim implicitly rests on the unstated premise that the chosen simulation suite faithfully represents real survey coadds.

axioms (1)
  • domain assumption The simulated PSF discontinuities and noise properties match those encountered in actual survey coaddition pipelines.
    Required for the conclusion that typical cases are unbiased to apply to real data.

pith-pipeline@v0.9.0 · 5683 in / 1255 out tokens · 29507 ms · 2026-05-21T22:29:06.376786+00:00 · methodology

Review history (2 revisions) →

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

  • IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear
    ?
    unclear

    Relation between the paper passage and the cited Recognition theorem.

    We investigate how discontinuities in the point spread function (PSF) and image noise affect weak gravitational lensing shear measurements... Using Metacalibration and Metadetection... we find no significant biases for typical galaxy populations, typical edge hit rates (a few percent), or coadds with tens to hundreds of input images.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages · 2 internal anchors

  1. [1]

    2021, The Astrophysical Journal Supplement Series, 253, 31, 10.3847/1538-4365/abd62c

    Abolfathi, B., et al. 2021, The Astrophysical Journal Supplement Series, 253, 31, 10.3847/1538-4365/abd62c

    work page doi:10.3847/1538-4365/abd62c 2021

  2. [2]

    2022, PASJ, 74, 247, doi: 10.1093/pasj/psab122

    Aihara , H., AlSayyad , Y., Ando , M., et al. 2022, , 74, 247, 10.1093/pasj/psab122

    work page doi:10.1093/pasj/psab122 2022

  3. [3]

    2024, arXiv e-prints, arXiv:2407.01771, 10.48550/arXiv.2407.01771

    Armstrong , R., Sheldon , E., Huff , E., et al. 2024, arXiv e-prints, arXiv:2407.01771, 10.48550/arXiv.2407.01771

    work page doi:10.48550/arxiv.2407.01771 2024

  4. [4]

    M., Armstrong , R., Krawiec , C., & March , M

    Bernstein , G. M., Armstrong , R., Krawiec , C., & March , M. C. 2016, , 459, 4467, 10.1093/mnras/stw879

    work page doi:10.1093/mnras/stw879 2016

  5. [5]

    1948, Annales d'Astrophysique, 11, 247

    de Vaucouleurs , G. 1948, Annales d'Astrophysique, 11, 247

    work page 1948

  6. [6]

    2025, , 697, A1, 10.1051/0004-6361/202450810

    Euclid Collaboration . 2025, , 697, A1, 10.1051/0004-6361/202450810

    work page doi:10.1051/0004-6361/202450810 2025

  7. [7]

    LSST: from Science Drivers to Reference Design and Anticipated Data Products

    Ivezi\'c, v., Tyson, J. A., Acosta, E., et al. 2008. 0805.2366v4

    work page internal anchor Pith review Pith/arXiv arXiv 2008

  8. [8]

    Kaiser , N. 2004. http://pan-starrs.ifa.hawaii.edu/project/people/kaiser/imageprocessing/im++.pdf

    work page 2004

  9. [9]

    2025, arXiv e-prints, arXiv:2506.16607, 10.48550/arXiv.2506.16607

    Li , X. 2025, arXiv e-prints, arXiv:2506.16607, 10.48550/arXiv.2506.16607

    work page doi:10.48550/arxiv.2506.16607 2025

  10. [10]

    LSST Science Book, Version 2.0

    LSST Science Collaboration . 2009, arXiv e-prints, arXiv:0912.0201, 10.48550/arXiv.0912.0201

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.0912.0201 2009

  11. [11]

    2018, , 56, 393, 10.1146/annurev-astro-081817-051928

    Mandelbaum , R. 2018, , 56, 393, 10.1146/annurev-astro-081817-051928

    work page doi:10.1146/annurev-astro-081817-051928 2018

  12. [12]

    2018, The LSST Dark Energy Science Collaboration (DESC) Science Requirements Document, arXiv, 10.48550/ARXIV.1809.01669

    Mandelbaum, R., et al. 2018, The LSST Dark Energy Science Collaboration (DESC) Science Requirements Document, arXiv, 10.48550/ARXIV.1809.01669

    work page doi:10.48550/arxiv.1809.01669 2018

  13. [13]

    Moffat , A. F. J. 1969, , 3, 455

    work page 1969

  14. [14]

    2025, arXiv e-prints, arXiv:2501.07938, 10.48550/arXiv.2501.07938

    Prat , J., & Bacon , D. 2025, arXiv e-prints, arXiv:2501.07938, 10.48550/arXiv.2501.07938

    work page doi:10.48550/arxiv.2501.07938 2025

  15. [15]

    P., & Burchat, P

    Sanchez, J., Mendoza, I., Kirkby, D. P., & Burchat, P. R. 2021, Journal of Cosmology and Astroparticle Physics, 2021, 043, 10.1088/1475-7516/2021/07/043

    work page doi:10.1088/1475-7516/2021/07/043 2021

  16. [16]

    2011, arXiv e-prints, arXiv:1109.6741, 10.48550/arXiv.1109.6741

    Sevilla , I., Armstrong , R., Bertin , E., et al. 2011, arXiv e-prints, arXiv:1109.6741, 10.48550/arXiv.1109.6741

    work page doi:10.48550/arxiv.1109.6741 2011

  17. [17]

    S., Becker , M

    Sheldon , E. S., Becker , M. R., Jarvis , M., Armstrong , R., & LSST Dark Energy Science Collaboration . 2023, The Open Journal of Astrophysics, 6, 17, 10.21105/astro.2303.03947

    work page doi:10.21105/astro.2303.03947 2023

  18. [18]

    S., Becker , M

    Sheldon , E. S., Becker , M. R., MacCrann , N., & Jarvis , M. 2020, , 902, 138, 10.3847/1538-4357/abb595

    work page doi:10.3847/1538-4357/abb595 2020

  19. [19]

    S., & Huff , E

    Sheldon , E. S., & Huff , E. M. 2017, , 841, 24, 10.3847/1538-4357/aa704b

    work page doi:10.3847/1538-4357/aa704b 2017

  20. [20]

    A., Long , H., Hirata , C

    Troxel , M. A., Long , H., Hirata , C. M., et al. 2021, , 501, 2044, 10.1093/mnras/staa3658

    work page doi:10.1093/mnras/staa3658 2021

  21. [21]

    R., Sheldon , E., et al

    Yamamoto , M., Becker , M. R., Sheldon , E., et al. 2025, arXiv e-prints, arXiv:2501.05665, 10.48550/arXiv.2501.05665

    work page doi:10.48550/arxiv.2501.05665 2025

  22. [22]

    Zackay , B., & Ofek , E. O. 2017, , 836, 187, 10.3847/1538-4357/836/2/187

    work page doi:10.3847/1538-4357/836/2/187 2017

  23. [23]

    Zernike , v. F. 1934, Physica, 1, 689, 10.1016/S0031-8914(34)80259-5

    work page doi:10.1016/s0031-8914(34)80259-5 1934