Lensing in the Blue III: Weak Lensing Shape Catalogs of 30 Merging Galaxy Clusters
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.
Vera: We're looking at the paper: "Lensing in the Blue III".
Jocelyn: Here's the gist of the paper. We present the weak gravitational lensing dataset from the Super-pressure Balloon-Borne Imaging Telescope (SuperBIT), which imaged 30 galaxy clusters during its 45 night flight in April to May 2023 The work constructs a calibrated shape catalog of these clusters, enabling unbiased reconstruction of the weak-lensing signal Data Acquisition and Preprocessing The SuperBIT mission utilized a modified Dall-Kirkham telescope to obtain near-diffraction-limited imaging from the stratosphere The instrument operated above 98% of the Earth’s atmosphere, allowing for near-space-quality imaging at a fraction of the cost of space missions The observations were conducted in three filters: F400W (u), F480W (b), and F600W (g), corresponding to ultraviolet, blue, and green bands Image pre-processing involved constructing master darks using "haircuts" instead of simple clipping to prevent biased mean dark current Hot pixels were identified and masked consistently throughout the pipeline, with background-subtracted images used for Bayesian shape measurement Shape Measurement Pipeline Precise measurement of individual galaxy shapes is achieved using the metacalibration algorithm, which measures the shear response of a shape estimator directly from data rather than relying on simulations This involves estimating the per-object response as Rγ = ∂e/∂γ at γ=0 The pipeline uses the ngmix package to perform a least-squares fit of a Gaussian mixture model to the PSF images, which are themselves interpolated models from PSFEx The final weighted-average shear estimator is given by ⟨γ⟩ ≈ ⟨Rγ⟩ − 1/⟨e⟩ − c total Pipeline Validation and Diagnostics The robustness of the pipeline is assessed through a comprehensive set of diagnostics, including assessments of point-spread function (PSF) modeling and tests for correlations between galaxy shapes and measured galaxy and PSF properties The authors analyze parallel sets of simulated images incorporating real observing conditions to yield a multiplicative shear bias of (1.1 ± 7.8) per cent They also examine correlations between the measured galaxy shear components and PSF properties, finding a statistically significant correlation between the mean galaxy shear ⟨e1⟩ and the PSF ellipticity ePSF1 with α1 = 0.140, corresponding to a 3.45 σ detection Catalog Construction and Performance Object selection for metacalibration is based on two critical criteria: the SNR of the galaxy model fit (SNR between 10 and 1000) and the ratio of intrinsic galaxy size to the PSF size (T/TPSF > 0.5) Weights are assigned to galaxies based on a two-dimensional parameter space of SNR and size ratio, defined as wij = ⟨Rγ⟩ squared / (σ 2e)ij The effective source number density is computed as neff = 1/A(P i wi) squared P i w 2i Simulation Results The simulation framework validates the pipeline by testing shear recovery against an NFW lens model, finding that the conventional calibration scheme yields a tangential shear bias of ∼ 6.3%, which is reduced to 1.
Title and authors: Vera: So what they’re summarizing here is the construction of this calibrated shape catalog for those thirty merging galaxy clusters. Essentially, they detail how they took the raw images from SuperBIT and processed them through their entire method to produce a catalog where the shapes are corrected, allowing for an unbiased reconstruction of the weak-lensing signal >
Jocelyn: They spend a lot of time laying out those preprocessing steps, like image stacking and source detection, before moving into the modeling of the point spread function. It’s a very thorough description of how they handle the observational non-idealities inherent in balloon-borne imaging >
Subrahmanyan: From my view, what's most interesting is their focus on incorporating real observing conditions directly into the calibration process via metacalibration, which is a way to naturally account for many observational non-idealities that simulations often miss >
Vera: Right. And they also define how they select objects for this catalog based on two criteria: the signal-to-noise ratio of the galaxy model fit being between one zero and one thousand, and the ratio of intrinsic galaxy size to the point spread function size being greater than zero point five >
Jocelyn: That weighting scheme is interesting because they assign weights based on a two dimensional parameter space of signal to noise ratio and that size ratio, defined as wij equals R gamma squared divided by sigma squared e ij >
Subrahmanyan: It seems like they are optimizing their sample selection not just for SNR but also for the physical relationship between the source size and the instrument's blurring effect, which is important when you're trying to understand baryonic components in these clusters >
Vera: And then they compute an effective source number density using that weight parameter space, where neff equals one over A of P i wij squared times P i w squared ij >
Jocelyn: So they’re essentially creating a statistically robust way to weigh each cluster in the final catalog based on how well it fits their criteria for accurate shape measurement >
Subrahmanyan: It shows they are thinking about the statistical power and how to maximize the constraints from this relatively small sample of thirty clusters when probing cosmology >
The paper's summary: Vera: Moving into the improvements, the paper suggests a few key things. First, they point out that using metacalibration allows them to obtain calibrated galaxy shapes for both their target clusters and even calibration fields, which is what enables unbiased reconstruction of the weak-lensing signal >
Jocelyn: They also highlight how they can use the ngmix package not just for fitting, but to perform a least-squares optimization of the posterior probability distribution for the galaxy model to get those best-fit galaxy parameters G >
Subrahmanyan: That parameter estimation step is crucial because it’s about extracting the actual physical properties of the galaxies from their observed shapes, which feeds directly into structure formation models >
Vera: And then they propose a novel calibration technique that’s based on a gridded shear response defined in the two dimensional space of metacalibration quantities, along with galaxy weights constructed in this parameter space to reduce the tangential shear bias from approximately six point three per cent down to one point one per cent >
Jocelyn: That reduction is what they're really emphasizing here, showing how this specific combination of gridded response and the new weighting scheme makes a tangible improvement in accuracy compared to older methods >
Subrahmanyan: That drop from six point three percent bias to one point one percent bias is substantial when you are trying to constrain cosmological parameters like sigma eight, because systematic errors have a direct impact on those measurements >
Vera: They also mention that the system can be designed to account for shear-dependent detection bias by implementing a metadetection framework, which they plan to use in future versions of the pipeline >
Jocelyn: So it’s not just about getting better at measuring shapes; it’s about building a whole framework that anticipates and mitigates other biases, like how the detection efficiency itself might change depending on the shear >
The paper's improvements: Vera: To wrap up, what this paper does is provide a comprehensive description of their SuperBIT data characteristics and preprocessing steps, detailing how they used metacalibration to build a calibrated shape catalog for thirty merging galaxy clusters >
Jocelyn: The key result is the validation of their pipeline against simulated images that incorporate real observing conditions, resulting in that one point one per cent multiplicative shear bias when tested against an NFW lens model >
Subrahmanyan: From the perspective of structure formation, this work gives us a statistically robust way to use these merging systems to study dark matter and baryonic interplay because it’s built on empirical calibration rather than purely simulated assumptions >
Vera: So, if you only listen for the big picture, this catalog is a step toward providing cleaner constraints on cosmology by minimizing those systematic errors that plague weak lensing measurements >
Jocelyn: And I think the next step is testing this calibrated catalog against larger datasets to see how it performs in real time, especially with these improvements they mentioned regarding detection bias >
Subrahmanyan: I just want to emphasize that the success of this calibration method hinges on how accurately they model those point spread functions, because if you mismodel the PSF, everything built on top of it gets skewed >
Vera: Right. And that’s what we have here for Lensing in the Blue III: Weak Lensing Shape Catalogs of thirty Merging Galaxy Clusters <ref:2603.18376#pg1,Lensing in the Blue III: Weak Lensing Shape Catalogs of 30 Merging>. We'll keep an eye on how these calibrated samples help us map out the large-scale structure of the universe and see what else these clusters tell us about dark energy >
Conclusion: Vera: So we've just walked through how they built their calibrated shape catalog from that SuperBIT data on thirty merging galaxy clusters >
Jocelyn: Yeah, they took those raw images and put them through a whole pipeline to get shapes that aren't as messy as they used to be >
Subrahmanyan: From a structure formation standpoint, the core of this paper is showing how you can use these specific mergers to map out dark matter distributions >
Vera: Right. And their main result here is that by using this metacalibration method with a gridded response, they managed to reduce the tangential shear bias from about six point three per cent down to one point one per cent >
Jocelyn: That drop is significant because those systematic errors are what we have to worry about when we try to measure things like the expansion rate of the universe >
Subrahmanyan: It tells us that this specific weighting scheme and calibration technique is quite robust for NFW lens models, which is a good sign for using these systems in cosmological constraints >
Vera: They also showed they can test their pipeline against real SuperBIT point spread functions from previous runs, which means they’re not just relying on simulations of the sky >
Jocelyn: And that's important because it shows they accounted for those observational non-idealities, like the way the atmosphere affects the image quality >
Subrahmanyan: So what this paper really does is prove that you can get a much cleaner signal from these telescopes even when you’re dealing with challenging conditions >
Vera: It gives us a solid catalog to work with for studying how massive structures grow in the universe right now >
Jocelyn: Yeah, it's a good piece of work, and we should keep an eye on how they use this calibration method on bigger surveys down the road >
Subrahmanyan: The next thing I'm interested in is seeing if we can apply this same kind of systematic reduction to other weak lensing surveys that look at larger numbers of clusters >
Vera: Exactly. And if we can keep these systematic errors under control, then this kind of data becomes really powerful for understanding the cosmic web >
Jocelyn: We'll see what the next set of observations brings, especially with how well these time-domain telescopes like Kepler are helping us now >
Subrahmanyan: Keep watching those large volume simulations; we need to know if this empirical calibration holds up when we move to even larger scales >
Vera: So that’s our rundown on Lensing in the Blue III: Weak Lensing Shape Catalogs of thirty Merging Galaxy Clusters >
Department of Physics, Northeastern University · Department of Physics, California Institute of Technology · Center for Data Science, New York University · Institute of Astronomy, University of Cambridge · Department of Applied Mathematics and Theoretical Physics, University of Cambridge · Kavli Institute for Cosmology, University of Cambridge · Institute for Computational Cosmology Department of Physics Durham University Centre for Extragalactic Astronomy Department of Physics Durham University Department of Physics Princeton University Institute for Aerospace Studies California Institute of Technology Centre for Advanced Instrumentation Durham University National Research Council Canada Herzberg Astronomy and Astrophysics Research Centre Scale AI Inc. Jet Propulsion Laboratory California Institute of Technology University of Toronto Institute for Aerospace Studies Department of Astronomy Steward Observatory University Arizona Laboratoire d’Astrophysique EPFL Observatoire de Sauverny StarSpec Technologies Inc. Imperial College London Department of Physics Case Western Reserve University David A Dunlap Dept of Astronomy and Astrophysics University Toronto Dunlap Institute for Astronomy and Astrophysics University Toronto Departamento de Física Teórica Universidad Autónoma de Madrid Institute of Physics National Yang Ming Chiao Tung University Cerro Tololo Inter-American Observatory NSF NOIRLab
astro-ph.CO, astro-ph.IM
Submitted: 2026-03-19
Updated: 2026-11-05
Comments: 35 pages, 27 figures, 5 table; matches with the published version in ApJ
Journal ref: Astrophys. J. 1011 (2026) 102
Code: https://github.com/GeorgeVassilakis/wcs
License: http://creativecommons.org/licenses/by/4.0/
Importance score: 79/100
The gist: We present the weak gravitational lensing dataset from the Super-pressure Balloon-Borne Imaging Telescope (SuperBIT)…
Key concepts
- Weak Gravitational Lensing
- This is an effect where massive objects like galaxy clusters bend the light from background galaxies, distorting their shapes. By measuring these tiny distortions across many galaxies, scientists can map the distribution of mass in space that is invisible through traditional means.
- Metacalibration Algorithm
- This technique measures how a shape estimator responds to shear directly from data without needing simulations. It estimates the per-object response (Rγ) at zero shear, allowing for an unbiased measurement of galaxy shapes, which is crucial for accurate weak lensing analysis.
- Tangential Shear Bias
- This refers to the systematic error in measuring the strength of gravitational distortion caused by clusters. The study shows that a novel calibration scheme reduces this bias significantly, from about 6.3% in conventional methods down to 1.1%, leading to more accurate mass estimates.
- SuperBIT Dataset
- This is the specific dataset consisting of imaging observations from the Super-pressure Balloon-Borne Imaging Telescope. It captured images of 30 galaxy clusters across three filters (UV, blue, green) during a single flight in 2023, providing the raw data for this lensing study.
Terminology
Summary
We present the weak gravitational lensing dataset from the Super-pressure Balloon-Borne Imaging Telescope (SuperBIT), which imaged 30 galaxy clusters during its 45 night flight in April to May 2023 The work constructs a calibrated shape catalog of these clusters, enabling unbiased reconstruction of the weak-lensing signal
Data Acquisition and Preprocessing
The SuperBIT mission utilized a modified Dall-Kirkham telescope to obtain near-diffraction-limited imaging from the stratosphere The instrument operated above 98% of the Earth’s atmosphere, allowing for near-space-quality imaging at a fraction of the cost of space missions The observations were conducted in three filters: F400W (u), F480W (b), and F600W (g), corresponding to ultraviolet, blue, and green bands Image pre-processing involved constructing master darks using haircuts
instead of simple clipping to prevent biased mean dark current Hot pixels were identified and masked consistently throughout the pipeline, with background-subtracted images used for Bayesian shape measurement
Shape Measurement Pipeline
Precise measurement of individual galaxy shapes is achieved using the metacalibration algorithm, which measures the shear response of a shape estimator directly from data rather than relying on simulations This involves estimating the per-object response as Rγ = ∂e/∂γ at γ=0 The pipeline uses the ngmix package to perform a least-squares fit of a Gaussian mixture model to the PSF images, which are themselves interpolated models from PSFEx The final weighted-average shear estimator is given by ⟨γ⟩ ≈ ⟨Rγ⟩ − 1/⟨e⟩ − c total
Pipeline Validation and Diagnostics
The robustness of the pipeline is assessed through a comprehensive set of diagnostics, including assessments of point-spread function (PSF) modeling and tests for correlations between galaxy shapes and measured galaxy and PSF properties The authors analyze parallel sets of simulated images incorporating real observing conditions to yield a multiplicative shear bias of (1.1 ± 7.8) per cent They also examine correlations between the measured galaxy shear components and PSF properties, finding a statistically significant correlation between the mean galaxy shear ⟨e1⟩ and the PSF ellipticity ePSF1 with α1 = 0. p. 1 al
residuals and their impact on weak-lensing measurements, as well as tests for correlations between
galaxy shapes and measured galaxy and PSF properties. To assess the multiplicative shear bias of the
pipeline, we analyze a parallel set of simulated images that incorporate the real observing conditions
from the flight, including measured SuperBIT PSFs, observed sky backgrounds, and detector noise,
yielding a bias of (1. We use the
metacalibration algorithm to obtain calibrated galaxy shapes for our target clusters and several
calibration fields, enabling unbiased reconstruction of the weak-lensing signal. We employ several
diagnostics throughout the pipeline, including assessments of point-spread function (PSF) modeling
Corresponding author: Sayan Saha
sa.Where the paper says it
Where the paper says it
Catalog Construction and Performance
Object selection for metacalibration is based on two critical criteria: the SNR of the galaxy model fit (SNR between 10 and 1000) and the ratio of intrinsic galaxy size to the PSF size (T/TPSF > 0.5) Weights are assigned to galaxies based on a two-dimensional parameter space of SNR and size ratio, defined as wij = ⟨Rγ⟩ squared / (σ 2e)ij The effective source number density is computed as neff = 1/A(P i wi) squared P i w 2i
Simulation Results
The simulation framework validates the pipeline by testing shear recovery against an NFW lens model, finding that the conventional calibration scheme yields a tangential shear bias of ∼ 6.3%, which is reduced to 1. p. 1 Consequently,
shear can only be inferred through statistical averaging of galaxy shapes under the assumption that the applied shear is locally constant. The intrinsic scatter in
galaxy ellipticities produces the so-called shape noise,
which constitutes the dominant statistical uncertainty
in weak-lensing measurements and can only be reduced
by increasing the source density.Where the paper says it
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Metacalibration algorithm implementation can be used to obtain calibrated galaxy shapes for target clusters and calibration fields, enabling unbiased reconstruction of weak-lensing signal.
-
The ngmix package can be used to perform a least-squares optimization of the posterior probability distribution for the galaxy model, obtaining the best-fit galaxy parameters G.
-
A novel calibration technique based on a gridded shear response defined in the two-dimensional space of metacalibration quantities, together with galaxy weights constructed in this parameter space, can be introduced to reduce tangential shear bias from approximately 6.3% to 1.1%.
-
The system can be designed to account for shear-dependent detection bias by implementing a metadetection framework, which plans to mitigate this effect in future versions of the pipeline.
-
The pipeline can be validated using empirical SuperBIT PSFs obtained from PSFEx runs on real data to faithfully capture observational non-idealities, and it can be tested against NFW cluster models located at the center of the field of view for shear recovery assessment.
Abstract
We present the weak gravitational lensing dataset from the Super-pressure Balloon-Borne Imaging Telescope (SuperBIT), which imaged 30 galaxy clusters during its 45 night flight in April to May 2023. SuperBIT is a first-of-its-kind balloon-borne imaging telescope that achieved near diffraction-limited observations in near-space conditions above 98% of the Earth's atmosphere. We use the metacalibration algorithm to obtain calibrated galaxy shapes for our target clusters and several calibration fields, enabling unbiased reconstruction of the weak-lensing signal. We employ several diagnostics throughout the pipeline, including assessments of point-spread function (PSF) modeling residuals and their impact on weak-lensing measurements, as well as tests for correlations between galaxy shapes and measured galaxy and PSF properties. To assess the multiplicative shear bias of the pipeline, we analyze a parallel set of simulated images that incorporate the real observing conditions from the flight, including measured SuperBIT PSFs, observed sky backgrounds, and detector noise, yielding a bias of (1.54 plus or minus 6.00) per cent.
Sources
- The Pan-STARRS1 Surveys
- Data Release 1 of the Dark Energy Spectroscopic Instrument
- Metacalibration: Direct Self-Calibration of Biases in Shear Measurement
- Large-Scale Cosmic Shear Measurements
- A Detection of Circumgalactic Dust at Megaparsec Scales with Maximum Likelihood Estimation
- GalSim: The modular galaxy image simulation toolkit
- Steps towards Nonlinear Cluster Inversion Through Gravitational Distortions. I. Basic Considerations and Circular Clusters
- Detection of correlated galaxy ellipticities on CFHT data: first evidence for gravitational lensing by large-scale structures
- Dark Energy Survey Year 6 Results: Cell-based Coadds and Metadetection Weak Lensing Shape Catalogue
Related papers
- 21cmEMUv3: a hybrid diffusion-LSTM emulator of 21cmFAST summary observables
- Connecting field-level and n-point function measurements of galaxy bias
- The Impact of Cosmic Variance and Satellites on JWST Clustering Measurements at Redshift around 6
- Universality of Primordial Anisotropies in Gravitational Wave Background
- KiDS-1000 cosmic shear reanalysis using MetaCalibration
- 2Flow2SN: Dusty Supernova Cosmology with Photometric Redshifts and Contamination using Neural Simulation-Based Inference