2608.25021v1
The ionizing properties of JWST's compact broad-line emitters
First listed 2026-08-27 | Last updated 2026-08-25
Abstract
The recent JWST discovery of a numerous population of broad-line emitters (BLEs) at $z>4$ has reopened the question of whether sources other than faint galaxies contributed significantly to cosmic reionization. To assess this contribution, we present a systematic census of compact, blue broad-line emitters at $4\leq z\leq7$ selected from the DJA, designed to isolate sources in which the UV continuum and the broad-line emission originate from the same compact physical region. Starting from a parent sample of 4145 galaxies with NIRSpec/PRISM spectroscopy and NIRCam imaging, we apply criteria on broad Ha emission (FWHM$>2000$km s-1, detected at S/N$\geq5$, complete at $L_{Ha,broad}\geq10^{42.7}$erg s-1), blue continuum slopes ($β_{UV}\leq-1.5$, $β_{opt}<0.5$), and morphological compactness in both the rest-frame UV and optical, returning a final sample of 20 sources. This constitutes $\sim20%$ of the PRISM-selected broad-line sample, which is dominated by Little Red Dots (LRDs). We find that the compact blue BLEs display hot ionizing continua, high-ionization UV lines (CIV, NIV], HeII), and elevated Lya emitter fractions ($X_{Lya}=50-67%$) relative to star-forming galaxies and LRDs. Beyond their similar lack of X-ray emission, their optical spectra closely resemble those of LRDs. This suggests that compact blue BLEs extend this population towards lower column density envelopes, but with similar engines, rather than representing a physically distinct population. Using Sirocco to model physical configurations that reproduce the observed spectra, we find a high ionizing photon production efficiency (log$ξ_{ion}\sim25.4$Hz erg-1) and a weighted average escape fraction $fesc\sim0.2$ for the bluest sources, and $\sim0.01$ for LRDs. This implies that broad-line sources do not dominate reionization globally, but given their luminosity, their contribution can dominate up to few Mpc scales.
Short digest
Mascia et al. build a systematically selected sample of 20 compact, blue broad-line emitters at 4≤z≤7 from 4,145 DJA galaxies with JWST/NIRSpec PRISM spectroscopy and NIRCam imaging, requiring luminous broad Hα, blue UV-to-optical slopes, and compact rest-UV and optical morphologies. These sources are only about 20% of the PRISM broad-line population, but show hot continua, CIV, NIV], and HeII emission, and Lyα-emitter fractions of 50–67%, while their optical spectra otherwise closely resemble Little Red Dots. Sirocco radiative-transfer models interpret the blue compact objects as lower-column-density viewing/covering configurations of the same underlying LRD-like engine, yielding log ξion≈25.4 and a weighted mean fesc≈0.2 for the bluest sources versus ≈0.01 for LRDs. The population is therefore unlikely to reionize the Universe globally, but its high luminosity can make it the dominant ionizing source within bubbles extending a few Mpc.
Key figures to inspect
- Figure 2. This is the essential selection figure: it places the compact blue BLEs in UV-versus-optical slope space, shows the βUV≤−1.5 and βopt<0.5 cuts, and visually distinguishes them from the broader BLE and star-forming-galaxy populations.
- Figure 9. The stacked-spectrum comparison directly supports the paper’s central population-level claim by showing that compact blue BLEs have distinctive high-ionization UV features while retaining optical spectral similarities to LRDs.
- Figure 11. This figure gives the physical mechanism behind the proposed blue-BLE–LRD connection: crossing from polar outflow to equatorial inflow increases the Balmer break and neutral column, while sharply suppressing LyC escape.
- Figure 13. The model-grid calibration turns an observable Balmer-break strength into an inferred LyC escape fraction, making it the key diagnostic bridge between the observed spectral sequence and the reionization argument.
- Figure 14. This synthesis figure presents the inferred ionizing-photon-production budget across star-forming galaxies, AGN, LRDs, and compact blue BLEs, directly framing the conclusion that broad-line populations are locally important but not global reionization dominants.
Discussion
Log in to view the paper discussion, see votes, and leave your own feedback.