[Paper Review] JADES: Physical properties of Ly$α$ and non-Ly$α$ emitters at z ~ 4.8-9.6

Nimisha Kumari, Renske Smit+22 more|arXiv (Cornell University)|Jun 17, 2024

Magnetism in coordination complexes6 citations

TL;DR

The study uses JWST/NIRSpec spectra from 253 galaxies in the JADES survey, stacking LAEs and non-LAEs across z~4.8–9.6 to derive dust, ionization, metallicity, and escape fractions, revealing distinct ISM conditions and challenges to applying local dust laws at high z.

ABSTRACT

We investigate the physical properties of Lyman-alpha emitters (LAEs) and non-Lyman-alpha emitters (non-LAEs) at z$\sim$4.8--9.6 via a stacking analysis of 253 JWST/NIRSpec spectra of galaxies observed as part of the JWST Advanced Deep Extragalactic Survey (JADES). We identify a sample of 42 LAEs with the equivalent width of Ly$α$ $\gtrsim$20Åand a sample of 211 non-LAEs, divide each sample further via the median redshift of the LAEs (z~6.3), and create composite spectra using the low and medium resolution spectra from NIRSpec. We estimate physical quantities such as dust extinction, UV continuum slope $β$, electron temperatures, ionization parameter, escape fraction of Ly$α$ and Lyman Continuum, and the photon production rate for each bin/stack. The existing dust-extinction laws do not appear to be valid at these epochs. The emission line ratio analyses show that active galactic nuclei might dominate all sub-samples, irrespective of Ly$α$ emission. LAEs show much higher [OIII]/[OII] and low [OII]/H$δ$ at z$\lesssim$6.3 compared to non-LAEs, but these line ratios are not sufficient to distinguish the two populations at z$>$6.3. However, the LAEs samples show large EW([OIII]4959, 5007) ($>$1000Å) compared to the non-LAEs sample at all redshifts. CIV/Ly$α$ and CIV/CIII] for LAE population at z$\lesssim$6.3 is $\sim$a factor of 5 larger than that for LAE population at z$>$6.3. The ionizing radiation for LAEs is hard, as revealed from several diagnostics, including CIV detection, high [OIII]/[OII] ($>$8), and large values of $ξ^{\star}_{ion}$.

Motivation & Objective

• Characterize the rest-frame UV–optical spectral properties of LAEs and non-LAEs across z~4.8–9.6.
• Compare physical ISM properties between LAEs and non-LAEs and across redshift bins.
• Infer dust content, ionization state, electron temperatures, metallicities, and escape fractions from composite spectra.
• Assess whether local dust-extinction laws apply to reionization-era galaxies and how Lyα presence correlates with ISM conditions.

Proposed method

• Assemble a sample of 253 robustly redshifted galaxies from JWST/JADES NIRSpec data spanning z~4.8–9.6.
• Identify LAEs as Lyα EW > 20 Å and split samples by LAE redshift median (z ~ 6.3).
• Create six composite spectra by de-redshifting, re-sampling, and normalizing R1000 spectra to the 1500 Å continuum, using weighted averages.
• Measure emission-line fluxes from Gaussian fits and correct for dust using Cardelli (1989) extinction, deriving E(B−V) from Balmer lines (Hγ/Hβ).
• Estimate Te([O III]) and Te([O II]) to derive direct gas-phase metallicities with PyNeb, and also compare with strong-line calibrations (O3, O2, R23, O32, Ne3O2).
• Compute ionization parameter log U from [O III]/[O II], and derive Lyα and LyC escape fractions using Lyα/Hβ ratios and β-based relations for LyC.

Experimental results

Research questions

• RQ1 What are the dust content, ionization state, and metallicity of LAEs versus non-LAEs at z ~ 4.8–9.6?
• RQ2 How do rest-frame UV–optical emission-line properties differentiate LAEs from non-LAEs across redshift bins?
• RQ3 Do high-redshift galaxies follow local or literature-extinction laws, and what do line ratios imply about the ionizing sources (star formation vs. AGN)?
• RQ4 How do Lyα and LyC escape fractions correlate with other physical properties (β, E(B−V), O32, etc.) across the samples?

Key findings

• LAEs show higher [O III]/[O II] and lower [O II]/Hδ at z ≲ 6.3 than non-LAEs, but these ratios are less diagnostic at z > 6.3.
• LAEs exhibit large EW([O III]4959,5007) (>1000 Å) across all redshifts compared to non-LAEs.
• Ionizing radiation in LAEs is hard, indicated by detections such as C IV, high [O III]/[O II] (>8), and large ξ*ion values.
• Dust-extinction laws used in the analysis do not appear universally valid at these epochs, and some LAE/non-LAE sub-samples imply AGN-dominated line ratios.
• C IV/Lyα and C IV/C III] in LAEs show redshift-dependent trends, with LAEs at z ≲ 6.3 having ~5× larger C IV/C III] than those at z > 6.3.
• The direct Te-based metallicities and indirect strong-line metallicities show notable scatter, likely due to differing ionization conditions; robust metallicity inferences require caution in high-z LAEs/non-LAEs.