First constraints from Ly-alpha forest limit DAO participation to at most 30% of dark matter for peaks below 50 h/Mpc at 95% CL using a deep kernel emulator of simulations.
Novel constraints on non-cold (non-thermal) Dark Matter from Lyman-$\alpha$ forest data
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
abstract
In this paper we present an efficient method for constraining both thermal and non-thermal Dark Matter (DM) scenarios with the Lyman-$\alpha$ forest, based on a simple and flexible parametrisation capable to reproduce the small scale clustering signal of a large set of non-cold DM (nCDM) models. We extract new limits on the fundamental DM properties, through an extensive analysis of the high resolution, high redshift data obtained by the MIKE/HIRES spectrographs. By using a large suite of hydrodynamical simulations, we determine constraints on both astrophysical, cosmological, and nCDM parameters by performing a full Monte Carlo Markov Chain (MCMC) analysis. We obtain a marginalised upper limit on the largest possible scale at which a power suppression induced by nearly any nCDM scenario can occur, i.e. $\alpha<0.03~{\rm{Mpc}}/h$ (2$\sigma$ C.L.). We explicitly describe how to test several of the most viable nCDM scenarios without the need to run any specific numerical simulations, due to the novel parametrisation proposed, and due to a new scheme that interpolates between the cosmological models explored. The shape of the linear matter power spectrum for standard thermal warm DM models appear to be in mild tension ($\sim 2\sigma$ C.L.) with the data, compared to non-thermal scenarios. We show that a DM fluid composed by both a warm (thermal) and a cold component is also in tension with the Lyman-$\alpha$ forest, at least for large $\alpha$ values. This is the first study that allows to probe the linear small scale shape of the DM power spectrum for a large set of nCDM models.
citation-role summary
background 1
citation-polarity summary
years
2026 6verdicts
UNVERDICTED 6roles
background 1polarities
background 1representative citing papers
Presents a tensorized GPU implementation of the 2-to-2 elastic self-collision operator for dark-sector particles and applies it to a two-source freeze-in scenario where self-interactions erase bimodal features.
An EFT-based field-level forward model for the Lyman-alpha forest matches simulations at the percent level on quasi-linear scales and generates mocks for DESI and DESI-II analyses.
Lyman-alpha forest data yield m_FDM > 1.9e-21 eV (95% CL) for pure FDM and f_FDM upper limits of 0.07-0.65 for mixed FDM at log10(m_FDM/eV) = -23 to -21.
Inelastic self-interacting dark matter with small mass splitting produces a cutoff in the matter power spectrum at k > 1 h Mpc^{-1} whose location depends on cross-section normalization, velocity dependence, dark matter mass and mass splitting, yielding non-monotonic exclusion regions from Lyman-α森林
A 1D convolutional neural network reconstructs the dark-matter phase-space distribution from the matter power spectrum with greater accuracy and broader applicability than an earlier empirical formula.
citing papers explorer
-
Strongest constraints on dark acoustic oscillations from the Lyman-alpha forest
First constraints from Ly-alpha forest limit DAO participation to at most 30% of dark matter for peaks below 50 h/Mpc at 95% CL using a deep kernel emulator of simulations.
-
Lyman-Alpha Forest and its Cross-Correlation with High-Redshift Galaxies in Effective Field Theory at the Field Level
An EFT-based field-level forward model for the Lyman-alpha forest matches simulations at the percent level on quasi-linear scales and generates mocks for DESI and DESI-II analyses.
-
Lyman-$\alpha$ forest constraints on pure and mixed fuzzy dark matter
Lyman-alpha forest data yield m_FDM > 1.9e-21 eV (95% CL) for pure FDM and f_FDM upper limits of 0.07-0.65 for mixed FDM at log10(m_FDM/eV) = -23 to -21.
-
Cosmology of Inelastic Self-Interacting Dark Matter: Linear Evolution and Observational Constraints
Inelastic self-interacting dark matter with small mass splitting produces a cutoff in the matter power spectrum at k > 1 h Mpc^{-1} whose location depends on cross-section normalization, velocity dependence, dark matter mass and mass splitting, yielding non-monotonic exclusion regions from Lyman-α森林
-
Machine Learning Does It and Does It Better: Unearthing Primordial Dark-Matter Velocities from the Matter Power Spectrum
A 1D convolutional neural network reconstructs the dark-matter phase-space distribution from the matter power spectrum with greater accuracy and broader applicability than an earlier empirical formula.