Mapping Cosmic Dawn: AI Unveils Early Universe’s Energy Secrets

Researchers from the University of Cambridge, the University of Trieste, and the Max Planck Institute for Astrophysics have developed a new method to map the epoch of reionization, a crucial period in the universe’s history when the first stars and galaxies illuminated the cosmos. This breakthrough could provide valuable insights into the energy industry’s understanding of the universe’s early energy dynamics.

The team, led by Soumak Maitra, has created a deep-learning framework called TORRCH, which stands for TOmographic Reconstruction of the Reionization of Cosmic Hydrogen. TORRCH uses the spatial distributions of Lyα emitters (LAEs) and non-Lyα-selected galaxies (NLSGs) to reconstruct the neutral-hydrogen fraction field during the epoch of reionization. This method moves beyond global summary statistics, such as the volume-averaged ionized fraction, to provide direct, field-level constraints on the ionization topology.

The researchers trained a deterministic 3D U-Net on mock surveys that span diverse reionization scenarios. They found that TORRCH could recover the large-scale ionization morphology from synthetic data comparable to current surveys with high fidelity. The method also reproduced both the one-point distribution and the 2D power spectrum of projected neutral fractions. Additionally, the predicted galaxy-IGM cross-correlation was captured well, including the expected small-scale anti-correlation and its decline towards zero at large separations.

The quality of the reconstruction depends on tracer completeness, with deep joint LAE+NLSG samples yielding the most accurate morphology. LAE-only selections retain bubble-scale topology but with reduced fidelity. The method is also robust to variations in ionization conditions between training and test data, and to realistic redshift uncertainties.

This research, published in the journal Monthly Notices of the Royal Astronomical Society, suggests that galaxy-based tomography can potentially deliver reliable reionization maps across realistic survey redshift windows. For the energy industry, understanding the epoch of reionization can provide insights into the early universe’s energy dynamics and the formation of the first stars and galaxies, which are crucial for understanding the origins of the elements that power stars and galaxies today.

This article is based on research available at arXiv.

Scroll to Top
×