Cosmological Constraints from Full-Scale Clustering and Galaxy-Galaxy Lensing with DESI DR1

TL;DR

This work develops a simulation-based framework to extract cosmological information from non-linear galaxy clustering and galaxy–galaxy lensing using DESI DR1 alongside DES, KiDS, and HSC lensing data. By employing AbacusSummit simulations and a decorated HOD with assembly bias, the authors fit multiple DESI samples (BGS2/3 and LRG1/2) down to small scales, and combine clustering with lensing to constrain $S_8$ and $ ext{Om}$ under Planck-like priors. The resulting constraints, $S_8 obreak= obreak 0.794 \, extpm\,0.023$ and $ ext{Om} obreak= obreak 0.295 \, extpm\,0.012$ (DES/KiDS) and $S_8 obreak= obreak 0.793 \,\pm\,0.017$ and $ ext{Om} obreak= obreak 0.303 \,\pm\,0.010$ (HSC, with fiducial photo-z offsets) favor a lower $S_8$ than Planck CMB values but remain broadly consistent within uncertainties, contributing to the ongoing discussion of the lensing-is-low and $S_8$-tension. The analysis highlights the constraining power of non-linear scales while acknowledging limitations from baryonic physics and photometric redshift calibrations, and points toward future improvements with DESI, deeper lensing data, and refined baryonic modeling.

Abstract

We present constraints on cosmic structure growth from the analysis of galaxy clustering and galaxy-galaxy lensing with galaxies from the Dark Energy Spectroscopic Instrument (DESI) Data Release 1. We analyze four samples drawn from the Bright Galaxy Survey (BGS) and the Luminous Red Galaxy (LRG) target classes. Projected galaxy clustering measurements from DESI are supplemented with lensing measurements from the Dark Energy Survey (DES), the Kilo-Degree Survey (KiDS), and the Hyper Suprime-Cam (HSC) survey around the same targets. Our method relies on a simulation-based modeling framework using the AbacusSummit simulations and a complex halo occupation distribution model that incorporates assembly bias. We analyze scales down to $0.4 \, h^{-1} \, \mathrm{Mpc}$ for clustering and $2.5 \, h^{-1} \, \mathrm{Mpc}$ for lensing, leading to stringent constraints on $S_8 = σ_8 \sqrt{Ω_\mathrm{m} / 0.3}$ and $Ω_\mathrm{m}$ when fixing other cosmological parameters to those preferred by the CMB. We find $S_8 = 0.794 \pm 0.023$ and $Ω_\mathrm{m} = 0.295 \pm 0.012$ when using lensing measurements from DES and KiDS. Similarly, for HSC, we find $S_8 = 0.793 \pm 0.017$ and $Ω_\mathrm{m} = 0.303 \pm 0.010$ when assuming the best-fit photometric redshift offset suggested by the HSC collaboration. Overall, our results are in good agreement with other results in the literature while continuing to highlight the constraining power of non-linear scales.

Figures (7)

• Figure 1: Best-fitting clustering and lensing predictions for the mock LRG1 measurements. The lower smaller panels indicate the difference between observations and model in units of the observational uncertainty. Gray regions indicate measurements not used in the fit.
• Figure 2: Mock constraints on $S_8$ and $\Omega_\mathrm{m}$. The different shaded regions indicate parameter spaces where $0 \leq \Delta \chi^2 < 2.28$ and $2.28 \leq \Delta \chi^2 < 5.99$ with respect to the best-fit cosmology, corresponding to the $68\%$ and $95\%$ confidence regions of a two-dimensional Gaussian distribution. The star indicates the values of $S_8$ and $\Omega_\mathrm{m}$ in the UNIT simulations. Finally, small grey dots indicate the cosmological parameters of AbacusSummit simulations used in this analysis.
• Figure 3: Similar to Fig \ref{['fig:mock_best_fit']} but for the DESI measurements with DES and KiDS lensing. The best-fit simulation, AbacusSummit_base_c103_ph000, is the one that minimizes the total $\chi^2$ of all three samples whereas the galaxy model is optimized individually for each sample.
• Figure 4: Similar to Fig \ref{['fig:data_best_fit']} but for the DESI measurements with HSC lensing amplitudes. The best-fitting simulation is AbacusSummit_base_c118_ph000.
• Figure 5: One-dimensional Bayesian posteriors on the galaxy model parameters for the four DESI samples with HSC lensing amplitudes under the best-fit cosmology. We also show the best-fit parameters as blue vertical lines. The different parameters are described in \ref{['tab:galaxy_parameters']}.