Modeling the Cosmological Lyman– α \alpha Forest at the Field Level

Roger de Belsunce, Mikhail M. Ivanov, James M. Sullivan, Kazuyuki Akitsu, Shi-Fan Chen · 2025

Read Modeling the Cosmological Lyman– α \alpha Forest at the Field Level by Roger de Belsunce, Mikhail M. Ivanov, James M. Sullivan, Kazuyuki Akitsu, Shi-Fan Chen (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-00284.txt.

Introduction

Roger de Belsunce and co-authors model the Lyman-alpha forest at the level of full fields, in a 2025 paper. The forest is a thicket of absorption lines in the light of distant quasars. It probes the young universe and the gas between galaxies at small scales. New surveys gather so much of it that sharper theory is needed to keep up.

Their analytic model grows the forest step by step from a given set of starting conditions. Against a leading fluid simulation, it matches key statistics to about percent accuracy down to a few million parsecs. Matches cover the flux spread, links with dark matter halos, and counts in cells. The authors say the tool bridges big-scale formulas and small-scale simulations. It arrives in time for a major sky survey's forest data.

Catalog record

CREATORRoger de Belsunce, Mikhail M. Ivanov, James M. Sullivan, Kazuyuki Akitsu, Shi-Fan Chen
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE39 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-00284

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Modeling the Cosmological Lyman– α \alpha Forest at the Field Level Preprint: MIT-CTP/5882 Roger de Belsunce, Mikhail M. Ivanov, James M. Sullivan, Kazuyuki Akitsu, Shi-Fan Chen Abstract The distribution of absorption lines in the spectra of distant quasars, called the Lyman- α \alpha (Ly- α \alpha ) forest, is a unique probe of cosmology and the intergalactic medium at high redshifts and small scales. The statistical power of ongoing redshift surveys demands precise theoretical tools to model the Ly- α \alpha forest. We address this challenge by developing an analytic, perturbative forward model to predict the Ly- α \alpha forest at the field level for a given set of cosmological initial conditions. Our model shows a remarkable performance when compared with the Sherwood hydrodynamic simulations: it reproduces the flux distribution, the Ly- α \alpha –dark matter halo cross-correlations, and the count-in-cell statistics at the percent level down to scales of a few Mpc. Our work provides crucial tools that bridge analytic modeling on large scales with simulations on small-scales, enabling field-level inference from Ly- α \alpha forest data and simulation-based priors for cosmological analyses. This is especially timely for realizing the full scientific potential of the Ly- α \alpha forest measurements by the Dark Energy Spectroscopic Instrument. Introduction. —Fluctuations in the cosmological density field encode information about the composition and evolution of the Universe. While galaxies, dark matter halos, and quasars trace the highly overdense regions, the low-density regions are filled with diffuse gas. In particular, neutral hydrogen in the low-density, highly ionized intergalactic medium (IGM) absorbs light, producing a distribution of absorption lines in observed quasar spectra, known as the Lyman- \alpha (Ly- \alpha ) forest. Measurements of these fluctuations in the neutral hydrogen density provide a powerful probe of the large-scale structure of the Universe in the high-redshift regime ( 2\lower 2.15277pt\hbox{$\;\mathrel{\mathop{\kern 0.0pt\sim}\limits^{<}}\;$}z\lower 2.15277pt\hbox{$\;\mathrel{\mathop{\kern 0.0pt\sim}\limits^{<}}\;$}4 ) as well as the thermal and ionization state of the IGM (see, e.g., McQuinn:2016 for a review). The large cross-section of the Ly- \alpha absorption of neutral hydrogen makes this observable sensitive down to Mpc distances and below, making it a unique source of cosmological information on small scales. In particular, the measurements of the line-of-sight power spectrum of the Ly- \alpha forest (see, e.g., Seljak:2005 ; Viel:2005 ; McDonald06 ; PYB13 ; Chabanier:2019 ; Pedersen:2020 ; 2023MNRAS.526.5118R ; 2024MNRAS.tmp..176K ) constrain neutrino physics Seljak:2005 ; Viel:2010 ; PYB13 ; Palanque2020 ; Ivanov:2024jtl ; He:2023oke ; He:2025jwp , primordial black holes Afshordi:2003 ; Murgia:2019 , dark matter Viel:2013 ; Baur:2016 ; Irsic17 ; Kobayashi:2017 ; Armengaud:2017 ; Murgia:2018 ; Garzilli:2019 ; Irsic:2020 ; Rogers:2022 ; Villasenor:2023 ; Irsic:2023 , the thermal state of the ionized IGM Zaldarriaga:2002 ; Meiksin:2009 ; McQuinn:2016 ; Viel:2006 ; Walther:2019 ; Bolton:2008 ; Garzilli:2012 ; Gaikwad:2019 ; Boera:2019 ; Gaikwad:2021 ; Wilson:2022 ; Villasenor:2022 , non-minimal cosmological models Goldstein:2023gnw ; Garny:2018byk ; Fuss:2022zyt , and the running of the spectral index Seljak:2006bg ; Ivanov:2024jtl . Combining multiple lines of sight allows one to reconstruct three-dimensional (3D) cosmological fluctuations on large scales. These 3D auto correlations of the forest and cross-correlations between the forest and quasars constrain the expansion history of our Universe through measurements of the baryon acoustic oscillations (BAO; McDonald:2007 ; Slosar2013 ; Busca:2013 ; dMdB:2020 ), and the broadband shape of the 3D correlation function Slosar2013 ; Cuceu:2021 ; Cuceu:2023 ; Gordon:2023 . The Ly- \alpha forest at high redshifts plays a crucial role in strengthening recent evidence for dynamical dark energy from the Dark Energy Spectroscopic Instrument (DESI) experiment DESI:2025zgx . DESI DESI:2016 ; DESI:2022 ; DESI_BAO_2024 ; DESI_lya_2024 will observe approximately one million quasar spectra with Ly- \alpha forest over its five-year observation period, producing the most precise map of our Universe at high redshift. Next-generation surveys such as DESI-II and Spec-S5 Schlegel:22_spec_roadmap ; Besuner25:spec_s5 will extract the Ly- \alpha forest from spectra of high-redshift galaxies, which will increase the total cosmological signal even further. These impressive experimental achievements call for robust theoretical modeling tools whose precision has to match statistical errors in order to convert the ongoing and upcoming Ly- \alpha measurements into precision measurements of cosmological parameters. Current Ly- \alpha forest modeling capabilities do not rise to the challenge posed by the incoming quantity and quality of this data. To date, all analyses of the three-dimensional Ly- \alpha forest data have been done at the level of the two-point function DESI_BAO_2024 (or its Fourier counterpart, the power spectrum Font-Ribera:2018 ; deBelsunce:2024knf ). While the two-point function analysis is computationally manageable and easy to interpret, the compression of a three-dimensional field into the two-point function is inevitably lossy Nguyen:2024yth ; Akitsu_FLI . A more optimal analysis may be a field level inference or the inclusion of three- and four-point function correlations. All these methods require consistent modeling of the Ly- \alpha forest beyond the two-point function. While such modeling is, in theory, possible by means of numerical hydrodynamic simulations, this method cannot be easily scaled to reproduce large cosmological volumes relevant to cosmic surveys. 1 1 1 For the analysis of large scales, different approximate prescriptions have been used to create large-volume Ly- \alpha forest mocks: AbacusSummit N -body simulations paint the forest on top of the dark matter field Hadzhiyska:2023 ; Hadzhiyska:2025cvk using a fluctuating Gunn-Peterson approximation (FGPA) calibrated on hydrodynamic simulations Croft98 ; 2022ApJ…930..109Q . More sophisticated techniques include the Ly- \alpha Mass Association Scheme (LyMAS; Peirani:2014 ; Peirani:2022 ), the Iteratively Matched Statistics method (IMS; Sorini:2016 ), Hydro-BAM ( 2022ApJ…927..230S ) , cosmic-web-dependent FGPA ( 2024A&A…682A..21S ) , and deep-learning reconstruction Jacobus:2024yev to connect the observed flux to the underlying matter field. In this Letter , we provide the theoretical tools for field-level modeling of the Ly- \alpha forest applicable to arbitrarily large volumes, enabling analysis modes and methodologies that have not been possible before. We develop a forward model for the Ly- \alpha forest flux fluctuations at the field level based on the framework of effective field theory (EFT; McDonald:2009dh ; Baumann:2010tm ; Carrasco:2013mua ; Ivanov:2022mrd ). The use of EFT is crucial for our purpose because it provides a systematic first-principles theoretical description of the Ly- \alpha field on large scales based on symmetries and dimensional analysis only. Conceptually, our model is similar to the field-level forward model for galaxies Schmittfull:2020trd ; Obuljen:2022cjo , but it includes the line-of-sight dependent operators dictated by symmetries of the Ly- \alpha forest McDonald:1999dt ; Givans:2020sez ; Desjacques:2018pfv ; Chen:2021rnb ; Ivanov:2023yla ; Ivanov:2024jtl ; Belsunce_Sullivan_skewspectrum . We test our model against the high-fidelity hydrodynamic Sherwood simulation Bolton:2016bfs , and find that it can accurately reproduce the simulated Ly- \alpha forest distribution on scales greater than a few Mpc. Crucially, in order to match the two realizations, the amplitudes and phases of all Fourier modes have to be correct. We show that our forward model can match all the amplitudes and phases of all Fourier modes, thus passing a more stringent accuracy test than comparing summary statistics which average over modes. Forward model. —The quantity of interest is the fluctuation field of the Ly- \alpha forest transmitted flux \delta_{F}=F/\overline{F}(z)-1 around the mean value of transmission \overline{F}(z) , which depends on redshift z . In what follows we aim to model the

Read the complete text