Sample variance denoising in cylindrical 21 cm power spectra
Open Access
- 13 November 2025
- journal article
- research article
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 703, A130
Abstract
State-of-the-art simulations of re-ionisation-era 21 cm signal have limited volumes, generally orders of magnitude smaller than observations. Consequently, the Fourier modes in common between simulation and observation have limited overlap, especially in cylindrical (2D) k -space that is natural for 21 cm interferometry. This makes sample variance (i.e. the deviation of the simulated sample from the population mean due to finite box size) a potential issue when interpreting upcoming 21 cm observations. Here, we introduce 21cmPSDenoiser , a score-based diffusion model that can be applied to a single, forward-modelled realisation of the 21 cm 2D power spectrum (PS), predicting the corresponding population mean on-the-fly during Bayesian inference. Individual samples of 2D Fourier amplitudes of wave modes relevant to current 21 cm observations can deviate from the mean by over 50% for 300 cmpc simulations, even when only considering stochasticity due to the sampling of Gaussian initial conditions. 21cmPSDenoiser reduces this deviation by an order of magnitude, outperforming current state-of-the-art sample variance mitigation techniques such as fixing and pairing by a factor of a few at almost no additional computational cost (∼2 s per PS). Unlike emulators, 21cmPSDenoiser is not tied to a particular model or simulator since its input is a (model-agnostic) realisation of the 2D 21 cm PS. Indeed, we confirm that 21cmPSDenoiser generalises to PSs produced with a different 21 cm simulator than those on which it was trained. To quantify the improvement in parameter recovery, we simulated a 21 cm PS detection by the Hydrogen Epoch of Reionization Arrays (HERA) and ran different inference pipelines corresponding to commonly used approximations. We find that using 21cmPSDenoiser in the inference pipeline outperforms other approaches, yielding an unbiased posterior that is 50% narrower in most inferred parameters.Keywords
This publication has 93 references indexed in Scilit:
- Percent-level timing of reionisation: Self-consistent, implicit-likelihood inference from XQR-30+ Lyα forest dataPublications of the Astronomical Society of Australia, 2025
- On the detection of a cosmic dawn signal in the radio backgroundNature Astronomy, 2022
- Array programming with NumPyNature, 2020
- SciPy 1.0: fundamental algorithms for scientific computing in PythonNature Methods, 2020
- LOFAR: The LOw-Frequency ARrayAstronomy & Astrophysics, 2013
- Reionization and the Cosmic Dawn with the Square Kilometre ArrayExperimental Astronomy, 2013
- Cosmology at low frequencies: The 21cm transition and the high-redshift UniversePhysics Reports, 2006
- The history of star formation in a cold dark matter universeMonthly Notices of the Royal Astronomical Society, 2003
- Transients from initial conditions: a perturbative analysisMonthly Notices of the Royal Astronomical Society, 1998
- Reverse-time diffusion equation modelsStochastic Processes and their Applications, 1982