Dark Energy Survey Year 6 Results: Cell-based Coadds and Metadetection Weak Lensing Shape Catalogue

We present the Metadetection weak lensing galaxy shape catalogue from the six-year Dark Energy Survey (DES Y6) imaging data. This dataset is the final release from DES, spanning 4422 deg$^2$ of the southern sky. We describe how the catalogue was constructed, including the two new major processing steps, cell-based image coaddition and shear measurements with Metadetection. The DES Y6 Metadetection weak lensing shape catalogue consists of 151,922,791 galaxies detected over riz bands, with an effective number density of $n_{\rm eff}$ =8.22 galaxies per arcmin$^2$ and shape noise of $σ_e$ = 0.29. We carry out a suite of validation tests on the catalogue, including testing for PSF leakage, testing for the impact of PSF modeling errors, and testing the correlation of the shear measurements with galaxy, PSF, and survey properties. In addition to demonstrating that our catalogue is robust for weak lensing science, we use the DES Y6 image simulation suite (Mau, Becker et al. 2025) to estimate the overall multiplicative shear bias of our shear measurement pipeline. We find no detectable multiplicative bias at the roughly half-percent level, with m = (3.4 $\pm$ 6.1) x $10^{-3}$, at 3$σ$ uncertainty. This is the first time both cell-based coaddition and Metadetection algorithms are applied to observational data, paving the way to the Stage-IV weak lensing surveys.

💡 Research Summary

This paper presents the construction, validation, and performance of the Metadetection weak‑lensing galaxy shape catalogue derived from the six‑year Dark Energy Survey (DES Y6) imaging data. The final DES Y6 footprint covers 4 422 deg² of the southern sky and contains 151,922,791 galaxies detected in the g, r, i, z bands. The catalogue achieves an effective number density of n_eff = 8.22 gal arcmin⁻² and a per‑galaxy shape noise of σ_e = 0.29, making it one of the deepest and most densely sampled weak‑lensing data sets to date.

Two major methodological innovations are introduced. First, the authors develop a “cell‑based coaddition” pipeline. Instead of creating a single, large coadded image, the survey area is tiled into ~1 arcmin cells. Within each cell, only those single‑epoch (SE) exposures that fully overlap the cell are combined, guaranteeing that no SE image edges intersect the cell interior. This eliminates the PSF discontinuities that arise in traditional coadds, allowing a well‑defined, spatially continuous PSF model for each cell. The trade‑off is a modest loss of depth because not all SE exposures contribute to every cell, but the gain in PSF fidelity is critical for precise shear measurement.

Second, the Metadetection (Metacalibration) algorithm is applied directly to the cell‑based coadds. Metadetection artificially shears each image in four directions (±γ₁, ±γ₂), runs the detection and shape measurement on each sheared version, and computes the shear response matrix R = ∂e/∂γ for every galaxy. By dividing the measured ellipticity by the response, the method self‑corrects for noise bias, model bias, and selection bias, including biases introduced by blending and detection thresholds. This approach is particularly well‑suited for the high‑blending regime expected in Stage‑IV surveys.

The data processing chain begins with DES DR2 products: calibrated SE images, Pixmappy astrometric solutions (including tree‑ring and chromatic distortion corrections), Piff PSF models that now incorporate colour‑dependent terms, FGCM photometric zero‑points, and associated weight maps, masks, and background images. The colour‑dependent PSF is propagated into the Metadetection step, ensuring that each galaxy’s effective PSF reflects its spectral energy distribution.

Extensive validation is performed in two complementary ways. Empirical null tests on the data examine PSF leakage (α), additive bias (c), and correlations of the shear with a suite of survey properties (airmass, seeing, sky background, stellar density, etc.). All tests show α ≲ 10⁻⁴ and c ≲ 10⁻⁴, well below the statistical uncertainties of the catalogue. To assess absolute calibration, the authors use a new DES‑Y6 image simulation suite (Mau et al. 2025). They generate 200 independent ΛCDM N‑body realizations (Ω_m = 0.26, σ₈ = 0.84, etc.) and construct 800 mock shear catalogues by sampling the simulated shear fields at the real galaxy positions and adding realistic shape noise. Running the full cell‑based + Metadetection pipeline on these simulations yields a multiplicative bias m = (3.4 ± 6.1) × 10⁻³, i.e., consistent with zero at the half‑percent level and comfortably satisfying the DES Y6 requirement of |m| < 1.3 %.

Compared with the previous DES Y3 Metacalibration catalogue, which exhibited a ∼1 σ bias due to blending, the new pipeline dramatically reduces both multiplicative and additive systematics. The authors also outline plans for an independent BFD‑based shear catalogue that will serve as a cross‑check, and they note ongoing improvements to the PSF colour model (Schutt et al. 2025).

In summary, the combination of cell‑based coaddition and Metadetection provides a robust, low‑systematics weak‑lensing shape catalogue for DES Y6, demonstrating that these techniques meet the stringent requirements of current analyses and are ready for deployment in upcoming Stage‑IV surveys such as LSST, Euclid, and Roman.