ELG$\times$LRG distribution through dark matter halo dynamics

TL;DR

We present HOMe, a physically motivated halo occupation framework, to generate high-fidelity DESI Y1 ELG, LRG, and ELG×LRG mock catalogs from AbacusSummit using a two-level Bayesian inference scheme. By incorporating intra-halo dynamics, halo exclusion, satellite velocity bias, and an environmental quenching mechanism via a joint-occupation condition, HOMe reproduces DESI two-point clustering down to 200 kpc scales with unprecedented precision. The analysis reveals that ELG satellites drive the small-scale anisotropic clustering, while most centrals are solitary and a substantial fraction are orphans for LRGs, with cross-tracer conformity emerging naturally from the forward model. The results constrain the satellite fractions, velocity biases, halo-exclusion parameters, and environmental quenching characteristics, providing a robust pathway to test cosmology and gravity with small-scale phase-space information. This framework enables high-fidelity, multi-tracer mocks that can inform covariance estimation and systematic studies for current and upcoming surveys.

Abstract

We investigate the clustering and halo occupation distribution (HOD) of DESI Y1 emission-line (ELGs) and luminous red (LRGs) galaxies at $0.8<z<1.1$, including their cross-correlation (ELG$\times$LRG), using the AbacusSummit suite and a new Halo Occupation Model (HOMe) for galaxy multi-tracers. This integrates intra-halo dynamics, halo exclusion, and quenching, bridging insights from hydrodynamical, HOD, abundance-matching, and semi-analytic studies. Leveraging full phase-space information from the Uchuu N-body simulation, and sampling satellites from dark-matter particle positions via physically motivated prescriptions, HOMe reproduces the anisotropic clustering down to $s=200\,h^{-1}$kpc with unprecedented accuracy. Model parameters are inferred solely from two-point statistics using a two-level Bayesian framework, yielding high-fidelity ELG, LRG and cross-reference catalogs. We find that satellite ELGs behave as incoherent flows within their parent halos, dominating the clustering below $4\,h^{-1}$Mpc. The HOD from the best-fit HOMe has the following properties: (i) 90.50% (85.91%) of ELGs (LRGs) are central galaxies without satellites, residing in halos of $M_{\rm vir}\sim6.6\times10^{11}\,(1.2\times10^{13})\,h^{-1}{\rm M}_\odot$; (ii) the ELG$\times$LRG cross-correlation is governed by central-central pairs and shaped by halo exclusion on $2-5\,h^{-1}$Mpc scales; (iii) 9.50% (14.09%) of ELGs (LRGs) are satellites, of which 1.09% (3.52%) inhabit halos with a central galaxy of the same species in a maximally conformal configuration, 7.02% (0.005%) orbit complementary hosts in a minimally conformal state, and 0.58% (10.57%) are orphans. HOMe high sensitivity precisely captures the dynamics of satellites in different host environments, opening a promising avenue for understanding systematics, the dynamical nature of dark matter, potentially distinguishing gravity models.

Figures (23)

• Figure 1: Radial profile of the AbacusSummit DM particles in hosts with typical ELG (blue triangles, dashed line) and LRG (red diamonds, solid line) halo virial masses, in units of $(h^{-1}{\rm M}_\odot)$. The error bars are the standard deviation in each radial bin.
• Figure 2: Normalised $R_{\rm vir}$ distribution of the AbacusSummit central host halos of ELG and LRG.
• Figure 3: Schematic illustration of the Home workflow. We start from the AbacusSummit input halo and DM particle catalogs and generate the latent satellite $V_{\rm peak}$ values in level-I inference. In level II, we apply the Home HOD prescription to assign satellite positions from DM particles, and the velocity bias model to generate their peculiar motions. Then, we perform galaxy-halo connection via abundance matching, probabilistically assigning ELG/LRG classes to the halos. Just prior to the matching, in the Home input catalog we model halo exclusion. We then mimic the environmental quenching of ELG satellites in massive LRG hosts statistically, through the joint occupation condition. Finally, by constraining Home physical parameters (37 in total) of our level-II inference, we obtain the high-fidelity mock catalog for DESI multi-tracers.
• Figure 4: $V_{\rm peak}$ functions (markers), and corresponding best fits (lines) based on Eq. \ref{['eq:klypincen']}, for the AbacusSummit host halos (black dots), compared to the Uchuu central (grey full squares almost perfectly overlapping with the black dots) and satellite (grey empty squares) halos at the fiducial redshift. For Uchuu the closest snapshot to $z_{\rm A}=0.8$ is $z_{\rm U}=0.78$. The uncertainties are computed from 30 bootstrap re-samplings. All the results are normalized to the simulation volume, that is $(2\,h^{-1}$Gpc)$^3$ for both AbacusSummit and Uchuu. The optimal parameters are reported in Table \ref{['tab:klypinpar']}. The posterior distributions are shown in Figure \ref{['fig:VFposterior']}
• Figure 5: Posterior distributions of the halo abundance fit in Eq. \ref{['eq:klypincen']} to the AbacusSummit hosts (top panel), Uchuu central (middle) and satellite (bottom) halos, as shown in Figure \ref{['fig:VFfits']}. The optimal parameters are reported in Table \ref{['tab:klypinpar']}.