Reading the Epic of Reionization


Editor’s note: Astrobites is a graduate-student-run organization that digests astrophysical literature for undergraduate students. As part of the partnership between the AAS and astrobites, we occasionally repost astrobites content here at AAS Nova. We hope you enjoy this post from astrobites; the original can be viewed at

Title: Constraining the neutral fraction of hydrogen in the IGM at redshift 7.5
Author: Austin Hoag, Maruša Bradač, Kuang-Han Huang, et al.
First Author’s Institution: University of California, Los Angeles
Status: Submitted to ApJ

The epoch of reionization (EoR) refers to a period in the universe’s history in which the element hydrogen, which to this day constitutes the majority of the baryonic matter in the universe, transitioned from being mostly neutral to mostly ionized (Figure 1). This transition occurred by a fairly intuitive process. Early on in the universe’s history, since there were no stars or galaxies to produce light, there were no energetic photons present to dislodge the electrons orbiting the nuclei of hydrogen atoms. However, once stars and galaxies began to form, there was an increase in the availability of such photons — and once the universe was roughly 1 billion years old, nearly 100% of hydrogen atoms had been ionized.

Figure 1: Timeline of the history of the universe, showing the EoR (click to enlarge). The term redshift in the x-axis label refers to the reddening of light from its rest color as a result of relative motion between the emitter and the observer. It is also written as “z“. [NAOJ]

Studying reionization presents an interesting puzzle for astronomers because we know the starting point (almost all hydrogen is neutral) and the ending point (almost all hydrogen is ionized), but have little idea of what path the universe took on its journey between the two. Did reionization start early or late in the universe’s history? Did it progress quickly or slowly? When did the various sources of ionizing photons (quasars, galaxy clusters, Population III stars, etc.) become abundant enough to have a significant effect on the transition?

One reason why these seemingly basic questions are still open is that astronomers mainly study light, and neutral hydrogen is notorious for absorbing lots of it. In fact, any photons with wavelengths of less than 912 Angstroms, the Lyman limit, will be absorbed by neutral hydrogen. Effectively, the neutral hydrogen that defines the EoR prevents it from being studied. This has forced astronomers to devise creative ways of studying this period of time.

Today’s paper covers one team’s strategy to measure the neutral hydrogen fraction, a value that is used to characterize the amount of remaining neutral hydrogen compared to the total. If this fraction equals one, then all the hydrogen in the universe is neutral. If it equals 0, then all the hydrogen in the universe is ionized. The larger the neutral hydrogen fraction, the deeper into the EoR measurements are probing.


The investigation detailed in today’s paper relied on a combination of observations and simulations of reionization. First, the authors observe the brightnesses of faint galaxies situated well within the EoR. Second, the authors use a simulation of reionization to model how the neutral fraction affects the observed brightness of galaxies. For the third and final step, they perform a statistical analysis comparing the observed and modeled brightnesses to find the neutral fraction at the redshift of their observed galaxies.

To obtain their sample of faint galaxies, the authors used MOSFIRE to observe several large galaxy clusters to target distant lensed background galaxies (Figure 2). A byproduct of general relativity, gravitational lenses are caused by massive foreground objects like galaxy clusters bending space. This bending effect focuses and amplifies the light of objects behind the lens, magnifying dim background galaxies and making them easier to detect. Singling out the lensed galaxy population in their data, the authors map out the distribution of brightnesses of these galaxies at a particular wavelength of light, called Lyman alpha (Lyα). Using Lyα is important both because it can be incredibly bright over large distances, and because the transmission of Lyα photons through space is strongly affected by the presence of neutral hydrogen.

Figure 2: One of the observations used in today’s paper, showing the contour of a modeled gravitational lens (orange line) and the locations of the authors’ targeted galaxies. [Hoag et al. 2019]

Using their observations and the modeled Lyα transmission from simulations, the authors use a Bayesian framework to infer the neutral hydrogen fraction contemporary with their observed Lyα-emitting galaxies. Inputting their observed Lyα measurements and corresponding galaxy ultraviolet luminosities into the model, they find that it is most likely that the neutral hydrogen fraction at a redshift of = 7.6 is about 0.88. In other words, when the universe was about 700 million years old, its hydrogen was about 88% neutral. This result is suggestive of a late and rapid reionization scenario, compared with some theories and simulations that have reionization approaching this neutral fraction at much higher redshifts.

Figure 3: The average hydrogen neutral fraction obtained in different studies; the legend indicates the method used to find the values. The results from today’s paper are indicated by the red star, located markedly above most of the other measurements. [Hoag et al. 2019]

The Epoch Conclusion

This finding is an especially important result given that the neutral fraction is strongly biased towards the environment that sources the observations, and most observations from the EoR are of unusually luminous objects. For example, previous observations of quasar spectra at similar redshifts have been suggestive of much lower neutral fractions (see e.g. “QSO damping wings” in Figure 3). Discrepancies like these could arise because the quasar itself is contributing ionizing photons towards reionization, decreasing the local neutral fraction. The authors’ choice to target faint galaxies, which are much more common than quasars, may be preferentially targeting a more representative region of space, thus giving a better estimate of the global neutral fraction.

There are a multitude of methods being used to measure the neutral hydrogen fraction during reionization, and every last one of them is crucial to understanding the early history of the universe. Today’s paper demonstrates the application of a powerful tool that can be added to the high-redshift astronomer’s arsenal.

About the author, Caitlin Doughty:

I am a fourth year graduate student at New Mexico State University. I use cosmological simulations to study galaxy evolution during the epoch of reionization, with a focus on metal absorption in the circumgalactic medium.

1 Comment

  1. Pingback: AAS Nova – New