Table of Contents
Fetching ...

Lyman-$α$ Visibility During the Epoch of Reionization: Combining JWST FRESCO Grism Data with Keck Archival Spectroscopy

Ecaterina Leonova, Rohan P. Naidu, Pascal A. Oesch, Gabriel Brammer, Jorryt Matthee, Romain A. Meyer, Daniel Schaerer, Mengyuan Xiao

TL;DR

This study targets Lyman-$\alpha$ visibility during the Epoch of Reionization ($z>7$) by linking JWST/FRESCO-detected rest-optical [OIII] emitters with archival Keck/MOSFIRE Ly$\alpha$ searches in GOODS-N and GOODS-S. By mapping seven groups of [OIII] emitters around bright LAEs and combining intrinsic-property analyses (via Prospector SED fits) with UV-LF-based overdensity estimates, the work assesses whether local overdensities or galaxy properties principally govern Ly$\alpha$ transmission. The authors recover eight new LAEs in GOODS-N and compile a 14-LAE sample when including literature sources, finding that Ly$\alpha$ visibility does not correlate simply with environmental overdensity; LAEs also show modestly higher [OIII] EW and bluer UV slopes, suggesting a multi-faceted origin involving dust, ionization, and potentially AGN or mergers. The results imply that large-scale environment, ionized-bubble size, viewing angle, and small-scale interactions collectively shape Ly$\alpha$ observability, highlighting the need for broader wavelength coverage and more complete sampling in future spectroscopic surveys to refine reionization models.

Abstract

The visibility of Lyman-$α$ emission at $z>7$ provides crucial insights into the reionization process and the role of galaxies in shaping the ionized intergalactic medium. Using JWST FRESCO data, we investigate the environments of Lyman-$α$ emitters (LAEs) in the GOODS-N and GOODS-S fields by identifying [OIII] emitters and analyzing their large-scale distribution. Using the FRESCO redshifts, we recover eight new LAEs from archival Keck/MOSFIRE observations at $z=7.0-7.7$, including a potential AGN candidate at $z \sim 7.2$. Complemented by six literature LAEs, our sample consists of 14 LAEs in total, all of which are [OIII] emitters except for one very faint source not detected by FRESCO. We define seven groups of [OIII] emitters centered around the brightest LAEs and find that these bright LAEs do not reside in more overdense environments than the average galaxy population. The overdensity parameters for LAEs and [OIII] emitters without Lyman-$α$, calculated for sources with $\mathrm{M_{UV}<-19.5}$ to ensure completeness, are similar, indicating that overdensities alone cannot fully explain Lyman-$α$ visibility. While LAEs have slightly higher recent star formation (SFR$_{10}$/SFR$_{50} \approx 1.3\times$) and [OIII] EW ($\approx1.5\times$), they show no significant differences from [OIII] emitters in UV slope ($β$), UV magnitude ($\mathrm{M_{UV}}$), or stellar mass ($\log_{\mathrm{M}_{\star}}$). Our results suggest that other factors may contribute to the observability of Lyman-$α$ emission. Future spectroscopic surveys with broader wavelength coverage and more complete sampling will be crucial for refining our understanding of reionization.

Lyman-$α$ Visibility During the Epoch of Reionization: Combining JWST FRESCO Grism Data with Keck Archival Spectroscopy

TL;DR

This study targets Lyman- visibility during the Epoch of Reionization () by linking JWST/FRESCO-detected rest-optical [OIII] emitters with archival Keck/MOSFIRE Ly searches in GOODS-N and GOODS-S. By mapping seven groups of [OIII] emitters around bright LAEs and combining intrinsic-property analyses (via Prospector SED fits) with UV-LF-based overdensity estimates, the work assesses whether local overdensities or galaxy properties principally govern Ly transmission. The authors recover eight new LAEs in GOODS-N and compile a 14-LAE sample when including literature sources, finding that Ly visibility does not correlate simply with environmental overdensity; LAEs also show modestly higher [OIII] EW and bluer UV slopes, suggesting a multi-faceted origin involving dust, ionization, and potentially AGN or mergers. The results imply that large-scale environment, ionized-bubble size, viewing angle, and small-scale interactions collectively shape Ly observability, highlighting the need for broader wavelength coverage and more complete sampling in future spectroscopic surveys to refine reionization models.

Abstract

The visibility of Lyman- emission at provides crucial insights into the reionization process and the role of galaxies in shaping the ionized intergalactic medium. Using JWST FRESCO data, we investigate the environments of Lyman- emitters (LAEs) in the GOODS-N and GOODS-S fields by identifying [OIII] emitters and analyzing their large-scale distribution. Using the FRESCO redshifts, we recover eight new LAEs from archival Keck/MOSFIRE observations at , including a potential AGN candidate at . Complemented by six literature LAEs, our sample consists of 14 LAEs in total, all of which are [OIII] emitters except for one very faint source not detected by FRESCO. We define seven groups of [OIII] emitters centered around the brightest LAEs and find that these bright LAEs do not reside in more overdense environments than the average galaxy population. The overdensity parameters for LAEs and [OIII] emitters without Lyman-, calculated for sources with to ensure completeness, are similar, indicating that overdensities alone cannot fully explain Lyman- visibility. While LAEs have slightly higher recent star formation (SFR/SFR) and [OIII] EW (), they show no significant differences from [OIII] emitters in UV slope (), UV magnitude (), or stellar mass (). Our results suggest that other factors may contribute to the observability of Lyman- emission. Future spectroscopic surveys with broader wavelength coverage and more complete sampling will be crucial for refining our understanding of reionization.

Paper Structure

This paper contains 13 sections, 4 equations, 13 figures, 8 tables.

Figures (13)

  • Figure 1: Spectra of Lyman-$\alpha$ emission for LAEs in the GOODS-N field. The top panel presents the 2D signal-to-noise spectra obtained using Keck MOSFIRE (Table \ref{['tab:LAE']} ), with each plot centered around the Lyman-$\alpha$ emission line. The dotted vertical line indicates the expected rest-frame Lyman-$\alpha$ position based on [OIII] redshift measurements. The bottom panel shows the corresponding 1D spectra, fitted with Gaussian profiles to model Lyman-$\alpha$ emission line.
  • Figure 2: The spatial distribution of Lyman-$\alpha$ and [OIII] emitters in GOODS-N is shown through 2D (top panel) and 3D (bottom panel) projections. The galaxies in the FRESCO field in GOODS-N are organized into four groups, each spanning 30 cMpc, with each group centered around the brightest LAE or one of the brightest galaxies in the region. Circles represent [OIII] emitters, stars indicate LAEs, and black squares highlight [OIII] emitters with spectra in the Keck archive at the Lyman alpha wavelength that were not detected as LAEs. The color scheme highlights the LAEs and their companions within each group: red for GN_z7_0, green for GN_z7_2, blue for GN_z7_5, and orange for GN_z7_6.
  • Figure 3: The spatial distribution of Lyman-$\alpha$ and [OIII] emitters in GOODS-S is shown through 2D (top panel) and 3D (bottom panel) projections. The galaxies in the FRESCO field in GOODS-S are organized into three groups, each spanning 30 cMpc, with each group centered around the brightest LAE or one of the brightest galaxies in the region. Circles represent the [OIII] emitters, while stars indicate the LAEs. The color scheme highlights the LAEs and their companions within each group: pink for GS_z7_2, yellow for GS_z7_7, and purple for GS_z7_8.
  • Figure 4: Left panel: The overdensities around the 7 central LAEs in each group within the GOODS-N and GOODS-S fields. Each curve is color-coded according to the same group scheme as in Figures \ref{['fig:GN2d']} and \ref{['fig:GS2d']}: red for GN_z7_0, green for GN_z7_2, blue for GN_z7_5, orange for GN_z7_6, pink for GS_z7_2, yellow for GS_z7_7, and purple for GS_z7_8. The overdensity parameter is calculated for sources with $\mathrm{M_{UV}<-19.5}$ to ensure completeness. Right panel: The mean overdensity for LAEs across all groups is represented by the blue curve, with the shaded region denoting the 16th to 84th percentile range. The gray line shows the overdensity for [OIII]-non LAEs in the GOODS-N and GOODS-S fields at $7 < z < 8$.
  • Figure 5: Properties of [OIII] emitters (green bars) and LAEs (purple bars): This comparison includes several key intrinsic properties, such as the [OIII] EW, $\mathrm{M_{UV}}$, $\beta$ slope, stellar mass ($\log_{\mathrm{M}_{\star}}$), log$\mathrm{sSFR_{10}}$, and SFR${10}$/SFR${50}$). The green and purple dashed lines represent the mean values for [OIII] and Lyman-$\alpha$ emitters, respectively. The numbers near each line indicate the mean value for each property, with error bars showing the 16th to 84th percentile range.
  • ...and 8 more figures