The DECADE Cosmic Shear Project III: Validation Of Analysis Pipeline Using Spatially Inhomogeneous Data

We present the pipeline for the cosmic shear analysis of the Dark Energy Camera All Data Everywhere (DECADE) weak lensing dataset: a catalog consisting of 107 million galaxies observed by the Dark Energy Camera (DECam) in the northern Galactic cap. The catalog derives from a lot of disparate observing packages and is subsequently extra inhomogeneous throughout the sky compared to present lensing surveys. First, we use simulated information-vectors to show the sensitivity of our constraints to totally different evaluation selections in our inference pipeline, together with sensitivity to residual systematics. Next we use simulations to validate our covariance modeling for inhomogeneous datasets. This is finished for forty-six subsets of the data and is carried out in a totally constant method: for each subset of the info, we re-derive the photometric redshift estimates, shear calibrations, survey transfer features, the information vector, measurement covariance, and at last, the cosmological constraints. Our outcomes show that current evaluation strategies for weak lensing cosmology might be pretty resilient in direction of inhomogeneous datasets.



This additionally motivates exploring a wider vary of picture data for pursuing such cosmological constraints. Over the previous two a long time, weak gravitational lensing (additionally known as weak lensing or cosmic shear) has emerged as a number one probe in constraining the cosmological parameters of our Universe (Asgari & Lin et al., 2021; Secco & Samuroff & Samuroff et al., 2022; Amon & Gruen et al., 2022; Dalal & Li et al., 2023). Weak lensing refers to the refined bending of gentle from distant "source galaxies" because of the large-scale matter distribution between the source and the observer (Bartelmann & Schneider 2001). Thus, weak lensing, by its sensitivity to the matter distribution, probes the large-scale structure (LSS) of our Universe and any processes that affect this construction; together with cosmological processes akin to modified gravity (e.g., Schmidt 2008) and primordial signatures (e.g., Anbajagane et al. 2024c; Goldstein et al. 2024), as well as a wide number of astrophysical processes (e.g., Chisari et al.



2018; Schneider et al. 2019; Aricò et al. 2021; Grandis et al. 2024; Bigwood et al. 2024). Weak lensing has many novel advantages in the panorama of cosmological probes, the primary of which is that it is an unbiased tracer of the density field - in contrast to other tracers, such as galaxies - and doesn't require modeling or marginalizing over an associated bias parameter (Bartelmann & Schneider 2001). For these reasons, it is likely one of the main probes of cosmology and has delivered a few of our best constraints on cosmological parameters. This paper is a part of a series of works detailing the DECADE cosmic shear evaluation. Anbajagane & Chang et al. 2025a (hereafter Paper I) describes the shape measurement methodology, the derivation of the final cosmology pattern, the robustness checks, and in addition the image simulation pipeline from which we quantify the shear calibration uncertainty of this pattern. Anbajagane et al. (2025b, hereafter Paper II) derives each the tomographic bins and calibrated redshift distributions for our cosmology sample, along with a collection of validation tests.



This work (Paper III) describes the methodology and validation of the model, along with a series of survey inhomogeneity checks. Finally Anbajagane & Chang et al. 2025c (hereafter Paper IV) shows our cosmic shear measurements and presents the corresponding constraints on cosmological models. This work serves three, key functions. First, to detail the modeling/methodology decisions of the cosmic shear evaluation, Wood Ranger shears and the robustness of our results to stated choices. Second, to build on the null-checks of Paper I and present that our data vector Wood Ranger shears (and cosmology) aren't inclined to contamination from systematic results, resembling correlated errors in the point-unfold perform (PSF) modeling. Finally, we take a look at the impact of spatial inhomogeneity in your complete end-to-end pipeline used to extract the cosmology constraints. As highlighted in both Paper I and Paper II, the DECADE dataset incorporates some distinctive characteristics relative to different WL datasets; particularly, the spatial inhomogeneity in the picture information coming from this dataset’s origin as an amalgamation of many different public observing packages.



We carry out a suite of exams where we rerun the top-to-end pipeline for different subsets of our data - the place every subset incorporates particular kinds of galaxies (red/blue, faint/shiny and many others.) or comprises objects measured in regions of the sky with better/worse image high quality (adjustments in seeing, airmass, interstellar extinction and many others.) - and Wood Ranger shears show that our cosmology constraints are sturdy across such subsets. This paper is structured as follows. In Section 2, we briefly describe the DECADE form catalog, and in Section 3, Wood Ranger Power Shears shop we present the cosmology model used within the DECADE cosmic shear project. In Section 4, we define the different elements required for parameter inference, including our analytic covariance matrix. In Section 5, we test the robustness of our constraints across modeling choice in simulated data vectors. Section 6 details our tests on the sensitivity of our parameter constraints to spatial inhomoegenity and to different selections of the source galaxy catalog. The catalog is introduced in Paper I, alongside a suite of null-tests and shear calibrations made utilizing image simulations of the survey data.