arXiv Science⌕ Search

arXiv · 2611.17425

Augmenting Galaxy Catalogues using Machine Learning in the FLAMINGO Simulations

Abstract

Hydrodynamic simulations are instrumental in the analysis of observations such as galaxy clustering used for the estimation of cosmological parameters. However, for these purposes, the simulations need both a large box size and high resolution, which is computationally prohibitive. In this work we use the connection between galaxies and their host dark matter halos to populate dark-matter-only simulations, using data from the FLAMINGO simulations suite. We employ the Extremely Randomized Trees algorithm to model the connection between the mass profile, history and environment of dark matter halos and the probability distribution of the stellar mass of the galaxy contained in the halo. We use halos matched between a dark-matter-only simulation and the corresponding hydrodynamical simulation to train a model for centrals. For satellites, we employ a Gaussian Processes emulator that simultaneously optimises the hyperparameters of a model and selects for the dark matter features that are predictive of the stellar mass while being insensitive to the differences between the hydrodynamic and dark-matter-only simulations, then train a model on only these features. We then apply these models to produce a new galaxy catalogue for the 2.8 Gpc FLAMINGO-10k dark-matter-only simulation as a product of this work. We demonstrate that this catalogue reproduces the statistics of the galaxy-halo connection predicted by the smaller volume hydrodynamical simulation on which the model was trained, including the stellar mass function, stellar mass-dark matter mass relation, 2-point correlation functions and correlations between stellar mass and dark matter concentration, half mass radius and dark matter mass within 3 Mpc.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Elliot Scott, Marika Asgari, Joop Schaye, Matthieu Schaller, Markus Michael Rau. 2026-11-12. Augmenting Galaxy Catalogues using Machine Learning in the FLAMINGO Simulations. https://arxiv.science/abs/2611.17425

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite arXiv Science by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related papers

Background dynamics and cosmological tests of interacting, viscous $H^2+H^{-2}$ dark energy

We study dark-sector energy exchange and bulk viscosity in the HHDE model, whose density combines terms proportional to the square and inverse square of the Hubble parameter. Five viscosity cases and seven interactions give 35 fluid realizations governed by a common autonomous system. When the intrinsic dark-energy pressure is reconstructed from the continuity equation, dark-energy viscosity leaves the expansion history unchanged. Interaction and dark-matter viscosity enter through their net contribution, producing further exact background degeneracies. Each realization admits stable de Sitter attractors in physical parameter domains, either dominated by dark energy or containing residual dark matter sustained by exchange and viscous stress. Representative posterior trajectories connect the inferred present states to these endpoints. Using supernovae, cosmic chronometers, DESI baryon acoustic oscillations and a compressed Planck distance prior, the density-product interaction yields a Hubble constant of $71.15^{+1.53}_{-1.64}$ km s$^{-1}$ Mpc$^{-1}$. Conditional on the applicability of that prior, the preferred transfer from dark energy to dark matter lowers the minimum chi-squared by 75.16 relative to noninteracting HHDE and reduces the discrepancy with the local distance-ladder measurement to approximately one standard deviation when both uncertainties are combined.

astro-ph.CO↗

Exploring the Dark Sector: Interacting Radiation in Light of Modern Cosmological Probes

[abridged] We constrain a phenomenological dark radiation (DR) framework consisting of free-streaming and fluid-like components, providing a model-independent extension of the standard radiation sector. Using Planck CMB data, DESI DR2 BAO measurements, and Pantheon+ and DES Y5 (Dovekie) supernova samples, we derive constraints on additional relativistic degrees of freedom and assess their impact on cosmological tensions. We obtain $N_{\rm fld}<0.66$ (95% C.L.) from CMB data alone, while the combination with BAO yields $N_{\rm fs}=2.93\pm0.23$ and $N_{\rm fld}=0.36^{+0.16}_{-0.21}$ (68% C.L.), consistent with Standard Model expectations for free-streaming radiation. The DR framework significantly alleviates the Hubble tension through an enhanced early-time expansion rate, which reduces the sound horizon scale. The tension with SH0ES is reduced from highly significant in $Λ$CDM to statistically non-significant for CMB+BAO data according to the $\mathcal{T}$-statistic. Bayesian model comparison shows no decisive preference for DR over $Λ$CDM when SH0ES is excluded, with results in the regime of weakly disfavoured. However, including SH0ES data leads to decisive Bayesian evidence in favour of the DR scenario. Overall, DR provides a compelling framework for resolving the Hubble tension. When CMB, BAO, Pantheon$+$ and SH0ES data are considered, we find an increased effective radiation content, $N_{\rm tot}=3.63^{+0.13}_{-0.15}$, with a fraction of free-streaming radiation, $f_{\rm fs}=0.392\pm 0.026$, a reduced sound horizon scale, $r_d = 141.8^{+1.3}_{-1.2}\,\mathrm{Mpc}$, and a higher primordial helium fraction, $Y_{\rm He}=0.2530 \pm 0.0017$, which lies at the level of approximately $\sim 2$-$2.5σ$ above direct determinations from metal-poor H II regions, while remaining broadly consistent with other abundance measurements within current uncertainties.

astro-ph.CO↗

Hermes - Towards an Optimal High-Performance Algorithm for Cosmic Statistics of Large Data Sets

We present Hermes, an in situ multiresolution framework for efficient and flexible measurements of cosmic large-scale-structure statistics from discrete catalogues. Hermes replaces conventional ex situ counting of particle tuples with algebraic operations among window-filtered continuous fields. A discrete point distribution is first reconstructed as a continuous density field in a compact scaling-function representation, in which two- and higher-point statistics are formulated as spatial products of locally filtered fields. This formulation expresses conventional binning for counts-in-cells and two- and higher-point correlation functions through window functions. Changing the window kernels defines generalised statistics without redesigning the estimator. We further introduce PyHermes, an open-source Python implementation of this framework, combining multiresolution field reconstruction with FFT-based window convolution, MPI/thread parallelism, and GPU acceleration for high-order field contractions. The framework supports a broad range of measurements, including isotropic and anisotropic two-point statistics, marked correlations, standard and multipole three-point functions, filtered statistics, and Fourier-space differential operators for constructing derived physical fields. Using halo catalogues from cosmological N-body simulations, we demonstrate PyHermes through a range of clustering statistics and use benchmarks to quantify its computational efficiency and scalability for large-scale data sets. By separating field representation from statistical windows, Hermes provides a unified computational framework in which a single reconstructed field can be reused for a wide range of standard and customised cosmic statistics. This field-window architecture enables scalable analysis of the large data sets expected from current and future galaxy surveys.

astro-ph.CO↗