Does the Escape Fraction Prescription Shape Reionization History and Morphology?
Abstract
The escape fraction of ionizing photons, fesc, is one of the least constrained quantities of the Epoch of Reionization (EoR), and the opacity of the intergalactic medium prevents its direct measurement at high redshifts. We constrain three power-law parameterizations, fesc(z), fesc(Mh), and fesc(z, Mh), against JWST UV luminosity functions and EoR observables (the ionizing emissivity, the neutral hydrogen fraction, and the Thomson optical depth) using Markov Chain Monte Carlo, with priors that allow fesc to either increase or decrease with redshift and halo mass. We then propagate the best-fit models into the semi-numerical simulation SimFast21 and compare their ionization morphology and 21 cm power spectra with those of a constant-fesc model. All models reproduce the observations with nearly identical global reionization histories. The data favor an escape fraction that decreases with increasing redshift, from ≈ 0.8 at z = 5 to ≈ 0.35 at z = 15. The halo-mass-only model prefers fesc increasing with halo mass, but this trend acts as a proxy for redshift evolution. When we vary both dependencies jointly, the data prefer a mass-independent fesc. At matched global neutral fraction, the ionization morphology is closely similar for all four prescriptions, and the 21 cm power spectra agree to within ≈ 10%. These results suggest that, once calibrated against the same observations, the details of the fesc prescription, which are set by the physics of the interstellar medium, largely average out in both the global observables and the 21 cm power spectrum. Distinguishing them may require statistics beyond the power spectrum, such as sightline statistics of the Lyα and 21 cm forests, or forecasts that exploit their small differences in reionization timing.
Related
- Probing the Reionization History and Bubble Sizes with JWST Lyman-α Fraction Measurements
- EDGE-INFERNO: The chemical signatures of assembly history in reionization fossils
- Constraining reionization-era Lyα escape with JELS-MUSE: a highly complete Hα-selected sample at z∼6.1
- EDGE-INFERNO: The chemical signatures of assembly history in reionization fossils
- Subaru meets JWST: A Direct Measurement of Lyα Escape Fraction at z≃6.2 with Dual Narrow-Band Imaging
BibTeX
@misc{sheoran2026does,
title = {{Does the Escape Fraction Prescription Shape Reionization History and Morphology?}},
author = {Pulkit Sheoran and Ankita Bera and Sultan Hassan},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.36046},
eprint = {2610.36046},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.36046},
}
RIS
TY - UNPB
TI - Does the Escape Fraction Prescription Shape Reionization History and Morphology?
AU - Pulkit Sheoran
AU - Ankita Bera
AU - Sultan Hassan
PY - 2026
DA - 2026/10/26
PB - arXiv
DO - 10.48550/arxiv.2610.36046
UR - https://arxiv.org/abs/2610.36046
AB - The escape fraction of ionizing photons, f_esc, is one of the least constrained quantities of the Epoch of Reionization (EoR), and the opacity of the intergalactic medium prevents its direct measurement at high redshifts. We constrain three power-law parameterizations, f_esc(z), f_esc(M_h), and f_esc(z, M_h), against JWST UV luminosity functions and EoR observables (the ionizing emissivity, the neutral hydrogen fraction, and the Thomson optical depth) using Markov Chain Monte Carlo, with priors that allow f_esc to either increase or decrease with redshift and halo mass. We then propagate the best-fit models into the semi-numerical simulation SimFast21 and compare their ionization morphology and 21 cm power spectra with those of a constant-f_esc model. All models reproduce the observations with nearly identical global reionization histories. The data favor an escape fraction that decreases with increasing redshift, from ≈ 0.8 at z = 5 to ≈ 0.35 at z = 15. The halo-mass-only model prefers f_esc increasing with halo mass, but this trend acts as a proxy for redshift evolution. When we vary both dependencies jointly, the data prefer a mass-independent f_esc. At matched global neutral fraction, the ionization morphology is closely similar for all four prescriptions, and the 21 cm power spectra agree to within ≈ 10%. These results suggest that, once calibrated against the same observations, the details of the f_esc prescription, which are set by the physics of the interstellar medium, largely average out in both the global observables and the 21 cm power spectrum. Distinguishing them may require statistics beyond the power spectrum, such as sightline statistics of the Lyα and 21 cm forests, or forecasts that exploit their small differences in reionization timing.
ER -