KiDS-1000 cosmic shear reanalysis using MetaCalibration
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.
Vera: I'm Vera.
Jocelyn: And I'm Jocelyn, here with Subrahmanyan.
Vera: We're looking at the paper: "KiDS-1000 cosmic shear reanalysis using MetaCalibration".
Jocelyn: Here's the gist of the paper. Detailed Research Synthesis: KiDS-1000 Cosmic Shear Reanalysis with MetaCalibration This research paper investigates potential systematic biases in cosmic shear measurements by re-analyzing data from the Kilo-Degree Survey (KiDS) using an alternative shape measurement pipeline, MetaCalibration, and comparing its performance against the established lensfit method.
Vera: The primary motivation stems from mild tensions observed between KiDS cosmic shear results and the Planck Cosmic Microwave Background (CMB) measurements.
Jocelyn: The core of the study lies in the application of MetaCalibration, a shape measurement pipeline designed to reduce biases inherent in traditional shear estimation methods.
Vera: The authors rigorously quantified these biases using multi-band image simulations, specifically utilizing SURFS-based KiDS-Legacy-Like Simulations (SKiLLS) to calibrate the results.
Paper summary: Vera: So we're looking at this paper titled "KiDS-one thousand cosmic shear reanalysis using MetaCalibration <ref:2510.01122#pg1,KiDS-1000 cosmic shear reanalysis using MetaCalibration>." Basically, the authors are taking some of the old KiDS data and running it through a new shape measurement pipeline called MetaCalibration to see if they can fix some biases that might be messing up the results when they compare them to things like the Planck measurements.
Jocelyn: Right, so it's about checking if there are systematic errors in how we measure those cosmic shears, and why that matters because those tensions between KiDS and Planck have been a bit confusing for a while.
Subrahmanyan: It sounds like they're digging into the input data quality again, which is always important when you’re trying to put cosmological constraints on top of it. What are they claiming about this approach?
Vera: They claim that by using MetaCalibration, specifically through these SURFS based KiDS Legacy Like Simulations called SKiLLS for calibration, they can significantly reduce the multiplicative bias in the shear measurements. They found that across all the different redshift bins, this multiplicative bias is less than zero point zero one seven.
Jocelyn: That sounds like a big reduction if it’s true, Vera. But what about the additive part of that bias? Is there anything else they found about how robust this method is when you look at the galaxies themselves?
Subrahmanyan: The paper suggests that this method is quite robust because it showed negligible sensitivity to key galaxy properties. That means even if you have some uncertainty in how we model a galaxy's shape or its spectral energy distribution, the overall bias correction doesn't change much. That’s a strong point for using it with data from surveys like KiDS-one thousand <ref:2510.01122#pg1>.
Vera: Exactly, it’s about being less reliant on perfect galaxy modeling. And they also looked at how well the Point Spread Function is modeled, and while imperfect PSF modeling can be an issue, they tried to handle that by making a stacked PSF model for each galaxy and only including chip images with at least thirty stars to keep things accurate.
Jocelyn: So it sounds like they’ve built a pretty solid pipeline here, trying to tackle those known issues in shear estimation. But what does this actually mean for the cosmological numbers we get? What are the final constraints they've derived?
Subrahmanyan: They used these corrected shear estimates to figure out constraints on cosmological parameters. The resulting inferred value for S eight is zero point seven eight nine plus or minus zero point zero two zero and minus or minus zero point zero two four.
Vera: That constraint, at S eight equal one, is consistent with what we got from the old KiDS lensfit results, but they say this new method improves the constraining power by about twenty-eight percent. That improvement comes from getting a higher effective source number density with MetaCalibration.
Jocelyn: So it’s not just getting a better measurement of one thing, it’s actually giving us tighter constraints on the universe itself because the input data is cleaner. But what about that tension they mentioned earlier with Planck?
Subrahmanyan: They conclude that the discrepancy with Planck isn't caused by errors in shear estimation at all. They say while the difference in S eight between their result and Planck is still around one point eight sigma, this gap has shrunk by half a sigma compared to previous studies. That strongly suggests the tension we see isn't due to systematic errors in how we measured the shear.
Vera: It’s encouraging because it means those historical discrepancies might not be artifacts of our measurement techniques. And they also confirmed that their recently finished KiDS Legacy analysis lines up with what Planck found earlier.
Jocelyn: So, if we take this paper, "KiDS-one thousand cosmic shear reanalysis using MetaCalibration," what's the big picture here <ref:2510.01122#pg1,KiDS-1000 cosmic shear reanalysis using MetaCalibration>? What's the real implication for how we interpret these large-scale structure measurements?
Subrahmanyan: It means that when you look at measurements of S eight from surveys like KiDS, you need to be careful about the pipeline you use. The authors show that a different approach can yield more reliable numbers and help us see if those tensions are real or just measurement noise.
Vera: It emphasizes that the input shapes and distances for the galaxies are a fundamental part of the puzzle when we try to connect these surveys to cosmology.
Jocelyn: So, we’ve seen how this reanalysis works and what it tells us about the systematic errors, which sets us up perfectly for looking at what these cosmological constraints actually imply for structure formation.
Conclusion: Vera: So we've seen how they used this MetaCalibration tool on KiDS data to clean up those shape measurements and get some tighter constraints on cosmic shear.
Jocelyn: Right, and now we’re looking at the title of this paper, "KiDS-one thousand cosmic shear reanalysis using MetaCalibration," and who actually wrote it.
Subrahmanyan: The authors are tackling that tension between KiDS measurements and what we see from the Planck Cosmic Microwave Background.
Vera: Basically, they're showing that this new way of measuring shapes doesn't introduce major systematic errors, which is a big deal for our cosmological models.
Jocelyn: They found this method significantly reduced the multiplicative bias in those measurements across all the different redshift bins.
Subrahmanyan: That reduction in bias means the resulting constraints on S eight are more reliable than what we had before.
Vera: It’s about getting a better handle on how much structure is out there, and whether that structure formation matches what our models predict.
Jocelyn: They also showed that this new pipeline can actually give us a larger source count, which means those S eight constraints get tighter by about twenty-eight percent.
Subrahmanyan: That’s interesting because it suggests the tension we've been seeing might not be due to bad measurement techniques.
Vera: Exactly, so this paper suggests that if we use better calibration tools like MetaCalibration, we can see if those discrepancies with Planck are real or just measurement noise.
Jocelyn: It’s a solid step in checking the reliability of these large-scale structure observations.
Subrahmanyan: It really forces us to consider the input data quality when we try to match different cosmological probes.
Leiden Observatory, Leiden University · Ruhr University Bochum, Faculty of Physics and Astronomy, Astronomical Institute (AIRUB) · Department of Physics, University of Oxford · Department of Physics and Astronomy, University College London · Institute for Astronomy, University of Edinburgh · School of Mathematics, Statistics and Physics, Newcastle University · Argelander-Institut für Astronomie, Universität Bonn
astro-ph.CO
Submitted: 2025-10-01
Updated: 2026-11-02
Comments: Published in A&A, 715, A11 (2026)
DOI: 10.1051/0004-6361/202557694
Code: https://github.com/rreischke/OneCovariance
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 100/100
The gist: KiDS-1000 Cosmic Shear Reanalysis with MetaCalibration This research paper investigates potential systematic biases in cosmic shear measurements by re-analyzing data from the Kilo-Degree Survey…
Key concepts
- MetaCalibration
- A new shape measurement pipeline designed specifically to reduce systematic biases inherent in traditional cosmic shear estimation methods. It was tested against established methods like lensfit and validated using simulations.
- Multiplicative Bias ($|m|$)
- A systematic error in shear measurements where the measured value is scaled by a constant factor. The analysis showed that MetaCalibration drastically reduced this bias, keeping it below 0.017, especially for distant sources.
- S8
- A cosmological parameter representing the amplitude of matter fluctuations in the universe. The study derived an S8 value of 0.789, which is consistent with previous results but provides a more precise constraint on cosmic structure.
Terminology
Summary
KiDS-1000 Cosmic Shear Reanalysis with MetaCalibration
This research paper investigates potential systematic biases in cosmic shear measurements by re-analyzing data from the Kilo-Degree Survey (KiDS) using an alternative shape measurement pipeline, MetaCalibration, and comparing its performance against the established lensfit method. The primary motivation stems from mild tensions observed between KiDS cosmic shear results and the Planck Cosmic Microwave Background (CMB) measurements. p. 0
Methodology and Bias Mitigation:
The core of the study lies in the application of MetaCalibration, a shape measurement pipeline designed to reduce biases inherent in traditional shear estimation methods. The authors rigorously quantified these biases using multi-band image simulations, specifically utilizing SURFS-based KiDS-Legacy-Like Simulations (SKiLLS) to calibrate the results.
Key findings regarding bias mitigation include:
-
Multiplicative Bias Reduction: The analysis demonstrates that the multiplicative bias (m) is significantly reduced, particularly for sources at more distant redshifts. Across all tomographic bins, the authors report a multiplicative bias of m < 0. p. 8017, with negligible additive bias. p. 0
-
Robustness to Galaxy Properties: A critical aspect of MetaCalibration's success is its robustness; the calibration process showed
negligible sensitivity to key galaxy properties,
suggesting it is less susceptible to uncertainties in galaxy modeling compared to other methods. -
PSF Modeling and Linearity: Further investigation into the measurement pipeline (as detailed by Yoon et al.) confirms the accuracy of the multiplicative bias estimation. p. 1 While imperfect Point Spread Function (PSF) modeling can introduce significant errors, the authors mitigated this by creating a stacked PSF model for each galaxy, excluding chip images with fewer than 30 stars to ensure accuracy. p. 1 Testing linearity showed that the multiplicative bias (m) does not depend on input shear up to gamma = 0. p. 7075; beyond this threshold, the relationship becomes non-linear, depending on the cube of the input shear. p. 6
Cosmological Constraints and Comparison with Planck:
The robust shear estimates derived from MetaCalibration were subsequently used to derive cosmological parameter constraints. The resulting inferred value for S 8 is 0. p. 3789+0.020-0.024. This constraint is consistent with previous KiDS-1000 lensfit results (S 8 = 1) but shows an improvement of approximately 28% in constraining power, attributed to the higher effective source number density achieved through MetaCalibration.
Crucially, the authors conclude that the tension with Planck is not attributable to shear measurement biases. p. 0 While the difference in S 8 between their result and Planck remains at a similar level (about 1. p. 08 sigma), this discrepancy has been reduced by 0. p. 05 sigma compared to previous studies, strongly suggesting that the observed tension is not caused by systematic errors in shear estimation. The analysis also confirms that the recently completed KiDS-Legacy analysis agrees with Planck results. p. 0
Sample Size and Final Assessment:
The MetaCalibration pipeline yielded a final sample size of 275,233,76 sources, which is 18% larger than the lensfit v2 catalogue and 30% larger than the lensfit v2 catalogue's source count in some appendices. Despite minor filtering steps (e. p. 2g., filtering out 56. p. 568% of objects due to blended sources), the overall conclusion is that MetaCalibration provides a more reliable shear calibration, and the resulting cosmological constraints are consistent with established results, reinforcing the conclusion that the tension observed between KiDS and Planck is likely not rooted in shear measurement artifacts. p. 0
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 shear estimation can be used as a main shear calibration process for large surveys like LSST, because
the calibration is less sensitive to the choice of galaxy population properties that are set to be inputs of the image simulation for shear calibration.
-
The system can produce cosmological parameter constraints with improved precision, finding that
the constraining power is improved by about 28%
compared to previous methods due to ahigher effective source density.
-
The AI system can provide robust estimates for multiplicative bias across different redshift bins, as the paper shows that
the residual multiplicative shear biases are small for all tomographic bins, with the largest bias mfinal = −0.0169 observed for the second bin.
-
The system can be made more robust against mismatches in galaxy properties by minimizing sensitivity to simulation inputs, because
MetaCalibration minimises the sensitivity to the image simulations.
-
The AI system can account for PSF anisotropy by convolving images with a kernel that circularises the PSF prior to shape measurement, as this approach is effective because
the resulting additive bias is small
and after applying an empirical correction,the additive bias is negligible.
-
The system can provide improved constraints on cosmological parameters by yielding tighter bounds on S 8 than previous analyses, as the MetaCalibration results show that the constraint
is improved by about 28% thanks to the increased effective source number density.
-
The system can perform a null test for residual systematics by computing COSEBIs B-mode null tests, which are expected to be consistent with zero, thus providing
a non-trivial test for residual systematics.
Sources
- The Wide Field Infrared Survey Telescope: 100 Hubbles for the 2020s
- Hyper Suprime-Cam Year 3 Results: Cosmology from Cosmic Shear Power Spectra
- Euclid preparation. LXVII. Deep learning true galaxy morphologies for weak lensing shear bias calibration
- Euclid. I. Overview of the Euclid mission
- Metacalibration: Direct Self-Calibration of Biases in Shear Measurement
- Weak lensing constraints on the stellar-to-halo mass relation of galaxy groups with simulation-informed scatter
- Hyper Suprime-Cam Year 3 Results: Cosmology from Cosmic Shear Two-point Correlation Functions
- KiDS-Legacy: Consistency of cosmic shear measurements and joint cosmological constraints with external probes
- KiDS-Legacy: Redshift distributions and their calibration
- KiDS-Legacy: Cosmological constraints from cosmic shear with the complete Kilo-Degree Survey
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
- 2Flow2SN: Dusty Supernova Cosmology with Photometric Redshifts and Contamination using Neural Simulation-Based Inference
- Angular clustering and bias of photometric quasars in the Kilo-Degree Survey Data Release 4