Ly$\alpha$ visibility from z = 4.5 to 11 in the UDS field: Evidence for a high neutral hydrogen fraction and small ionized bubbles at z $\sim$ 7

This study utilizes JWST and ground-based spectroscopy of 651 galaxies in the UDS field to trace Ly$\alpha$ evolution from z=4.5 to 11, revealing a significant drop in Ly$\alpha$ visibility at z$\sim$7 that implies a high neutral hydrogen fraction and the presence of small, patchy ionized bubbles, thereby highlighting the inhomogeneous nature of cosmic reionization.

The Gist

Imagine the early universe as a giant, thick fog. For the first few hundred million years after the Big Bang, this fog was made of neutral hydrogen gas. It was so thick that light, specifically a specific type of ultraviolet light called Lyman-alpha (Lyα), couldn't travel through it. It was like trying to shout across a dense forest; the sound (light) would get scattered and absorbed before it could reach anyone.

Then, the first stars and galaxies began to form. They acted like giant heaters, burning through the fog and creating clear pockets of air. This process is called Reionization. Over time, these clear pockets grew and merged until the entire universe became transparent, allowing light to travel freely.

This paper is a detective story about when and how that fog cleared up, specifically looking at a patch of sky called the UDS field.

Here is the breakdown of their findings using simple analogies:

1. The "Flashlight" Test

The astronomers used the James Webb Space Telescope (JWST) as a giant flashlight. They looked at 651 ancient galaxies (some as old as 13 billion years) to see if their "Lyα flashlights" could be seen.

• The Goal: If the galaxy's light gets through, the fog (neutral hydrogen) in that direction is thin or gone. If the light is blocked, the fog is still thick.
• The Result: They found that between redshift 5 and 6 (a time when the universe was already mostly clear), the light was easy to see. But at redshift 7 (a slightly earlier time), the light suddenly started getting blocked in the UDS field.

2. The "Slit" Problem (Why the numbers looked weird)

The researchers noticed something strange. When they compared their JWST data to older ground-based telescope data, the JWST numbers were lower. It looked like fewer galaxies were emitting light than expected.

• The Analogy: Imagine you are trying to measure the brightness of a streetlamp.
  • Ground telescopes use a wide net (a wide slit) to catch the light.
  • JWST uses a very narrow straw (a tiny slit).
  • Because Lyα light scatters and spreads out like a puff of smoke around the galaxy, a lot of it misses the narrow straw.
• The Fix: The team realized JWST was "losing" about 35% of the light simply because the straw was too narrow. Once they corrected for this, the data from JWST and the ground telescopes finally matched up.

3. The "Patchy Fog" Discovery

This is the most exciting part. The universe didn't clear up like a curtain being pulled back all at once. It cleared up like a patchy fog on a cold morning.

• The EGS Field (The Sunny Spot): In a different part of the sky (the EGS field), the fog was almost completely gone at redshift 7. It was like a sunny day; the light traveled freely.
• The UDS Field (The Foggy Spot): In the UDS field, the fog was still very thick. The researchers calculated that 70% to 90% of the hydrogen was still neutral (foggy) in this specific direction.
• The Conclusion: Reionization wasn't a uniform event. Some places cleared up early, while others were still stuck in the fog. It was a messy, uneven process.

4. The "Islands of Light"

Even in the thick fog of the UDS field, the astronomers found two tiny "islands" where the light could get through.

• The Analogy: Imagine you are in a dense forest, but you see two small clearings where the trees have been cut down.
• The Findings: They found two specific groups of galaxies (at redshifts 7.29 and 7.77) that had cleared out their own little bubbles of space. These bubbles were about 0.5 to 0.6 million light-years across.
• Why it matters: These bubbles are the "seeds" of the clear universe we live in today. They show us that galaxies had to work together in clusters to burn away the fog locally before the whole universe could clear up.

Summary

This paper tells us that:

1. JWST is great, but we have to be careful: Its tiny "straw" misses some scattered light, so we need to do math to fix the numbers.
2. The universe cleared up unevenly: At the same time in history, one part of the sky was bright and clear, while another part was still thick with fog.
3. Galaxies are the heroes: They didn't just wait for the fog to lift; they actively burned holes in it, creating small bubbles that eventually merged to make the universe transparent.

It's a story of a universe waking up, but not all at once—some neighborhoods woke up early, while others were still hitting the snooze button.

Technical Summary

Here is a detailed technical summary of the paper "Lyα visibility from z = 4.5 to 11 in the UDS field: Evidence for a high neutral hydrogen fraction and small ionized bubbles at z ∼7".

1. Problem Statement

The Epoch of Reionization (EoR) marks the transition of the intergalactic medium (IGM) from a neutral to an ionized state. A primary goal of modern cosmology is to trace the volume-averaged neutral hydrogen fraction ($X_{\text{HI}}$) and understand the spatial inhomogeneity of this process.

• The Probe: Lyman-$\alpha$ (Ly$\alpha$) emitters (LAEs) are robust tracers of the IGM because Ly$\alpha$ photons are resonantly scattered by neutral hydrogen. A drop in the fraction of galaxies showing Ly$\alpha$ emission ($X_{\text{Ly}\alpha}$) at high redshift ($z > 6$) indicates increasing IGM neutrality.
• The Tension: Previous ground-based spectroscopic surveys suggested a significant drop in $X_{\text{Ly}\alpha}$ at $z > 6$. However, early James Webb Space Telescope (JWST) results showed conflicting trends, with some fields (like EGS) showing high Ly$\alpha$ visibility even at $z \sim 7$, while others remained ambiguous.
• The Gap: There was a lack of systematic, spectroscopically confirmed Ly$\alpha$ data in the Ultra Deep Survey (UDS) field to compare with other JWST fields (EGS, GOODS) and ground-based surveys. Additionally, discrepancies between JWST and ground-based Ly$\alpha$ measurements at $z \sim 6$ (where the IGM should be fully ionized) needed resolution to ensure accurate $X_{\text{HI}}$ constraints.

2. Methodology

The authors conducted a comprehensive analysis of the UDS field using a combined sample of spectroscopic data.

• Data Compilation:
  • Sample: 651 spectroscopically confirmed galaxies at $4.5 \le z \le 11$.
  • Sources: 135 from the CAPERS survey (JWST-NIRSpec PRISM) and 516 from the DAWN JWST Archive (DJA) (including RUBIES, GTO-1215, etc.).
  • Spectroscopy: 531 spectra from NIRSpec-PRISM ($R \sim 30-300$) and 120 from medium-resolution modes ($R \sim 1000$).
• Data Processing & Analysis:
  • Redshifts: Secure spectroscopic redshifts ($z > 4.5$) were established via visual inspection of UV/optical lines and Ly$\alpha$ breaks.
  • AGN Removal: Active Galactic Nuclei (AGNs) were identified and excluded using broad Balmer line fitting and Bayesian Information Criterion (BIC) analysis. 24 broad-line AGNs were removed.
  • Ly$\alpha$ Measurement: A forward-modeling approach was used to fit Ly$\alpha$ emission lines, accounting for instrumental resolution and IGM absorption on the blue side.
  • Slit Loss Correction: To address the discrepancy between JWST and ground-based data, the authors compared the cumulative distribution functions (CDFs) of Ly$\alpha$ Equivalent Width ($EW_0$) between JWST data and the ground-based De Barros et al. (2017) sample at $z \sim 6$.
• IGM Modeling:
  • The observed $EW_0$ distributions at $z=7$ were compared against theoretical transmission models (Dijkstra et al. 2011) to infer $X_{\text{HI}}$.
  • Ionized Bubbles: An iterative algorithm identified spectroscopic companions around Ly$\alpha$ emitters to estimate the size of ionized bubbles ($R_{\text{ion}}$) based on the ionizing photon budget ($\dot{N}_{\text{ion}}$) and observed physical distances.

3. Key Contributions & Results

A. Resolution of JWST vs. Ground-Based Discrepancy

• Observation: At $z \sim 6$, JWST measurements of $X_{\text{Ly}\alpha}$ were consistently lower than ground-based results (e.g., VLT-FORS2), despite the IGM being fully ionized.
• Cause: The authors determined this is an observational bias due to Ly$\alpha$ slit loss.
• Quantification: By comparing CDFs, they found a transmission factor of $T = 0.65 \pm 0.10$, implying a **slit loss of $35 \pm 10%$** in JWST spectra. This is attributed to the resonant nature of Ly$\alpha$scattering, which extends the emission spatially beyond the narrow JWST slit ($0.2''$) compared to wider ground-based slits ($\sim 1''$).

B. Evolution of Ly$\alpha$ Visibility ($X_{\text{Ly}\alpha}$)

• $z = 5-6$: The UDS results align with the average of other JWST fields, showing $X_{\text{Ly}\alpha} \approx 15-21\%$. This confirms the IGM is largely ionized.
• $z = 7$: A significant field-to-field variation was observed:
  • EGS Field: High $X_{\text{Ly}\alpha}$ (anomalously high), suggesting a large ionized bubble.
  • UDS Field: Low $X_{\text{Ly}\alpha}$ ($5.0^{+6}_{-1.9}%$), indicating a largely neutral IGM.
• $z = 8-9$: Data is limited by small-number statistics, but shows large scatter.

C. Constraints on Neutral Hydrogen Fraction ($X_{\text{HI}}$)

• Using the $z=6$ DB17 sample as a reference for a fully ionized universe, the authors modeled the attenuation at $z=7$.
• UDS Field: The low Ly$\alpha$ visibility implies a high neutral fraction of $X_{\text{HI}} = 0.7 - 0.9$.
• EGS Field: Consistent with a fully ionized IGM ($X_{\text{HI}} \approx 0$).
• Conclusion: The reionization process is highly patchy and inhomogeneous, with the UDS and EGS fields representing opposite extremes of the IGM state at $z \sim 7$.

D. Identification of Ionized Bubbles

• The authors identified two robust ionized bubbles in the UDS field where Ly$\alpha$ photons could escape despite the high global $X_{\text{HI}}$:
  1. $z = 7.29$ (RUBIES-22104): Radius $R_{\text{ion}} = 0.6$ pMpc; Photometric overdensity $N/\langle N \rangle = 3$.
  2. $z = 7.77$ (CAPERS-142615): Radius $R_{\text{ion}} = 0.5$ pMpc; Photometric overdensity $N/\langle N \rangle = 4$.
• These bubbles are significantly smaller than the massive ($2-12$ pMpc) bubbles found in the EGS field, consistent with a scenario where the UDS field is dominated by neutral gas with only small, localized ionized pockets.

4. Significance

• Instrumental Calibration: The paper provides a critical correction factor ($\sim 35\%$ slit loss) for JWST Ly$\alpha$ measurements, essential for reconciling space-based and ground-based data and accurately measuring $X_{\text{HI}}$.
• Reionization Topology: It provides the first direct evidence of extreme field-to-field variance at $z \sim 7$ within the same survey depth, confirming that reionization was not a uniform process but a "patchy" one driven by local overdensities.
• Physical Constraints: The identification of small ionized bubbles in a high-$X_{\text{HI}}$ environment supports hydrodynamical simulations predicting that in highly neutral IGMs, only small, dense clusters of galaxies can create sufficient ionizing photons to clear local bubbles.
• Future Directions: The authors emphasize the need for deeper JWST-NIRSpec observations (to reach $EW_0 \sim 10$ Å) and the use of NIRSpec-IFU (Integral Field Unit) to create a slit-loss-free reference sample for future high-redshift studies.

In summary, this work establishes the UDS field as a region of high neutral hydrogen at $z \sim 7$, contrasting sharply with the EGS field, and quantifies the instrumental biases necessary to interpret JWST Ly$\alpha$ data correctly.