Constraints on the Gas Geometry Surrounding Little Red Dots through Narrow-Line Diagnostics

Visal Sok, Erica J. Nelson, Mitchell C. Begelman, Jason Dexter, Francesco D'Eugenio, Jenny E. Greene, Joel Leja, Katherine E. Whitaker, Andrew J. Bunker, Pablo G. Pérez-González, Pierluigi Rinaldi, Alberto Torralba, Hannah Übler

First listed 2026-06-24 | Last updated 2026-06-22

Abstract

Little Red Dots (LRDs) are a recently identified population of high-redshift sources, with a common interpretation being accreting black holes embedded within a spherical, optically thick gas envelope. Within this framework, some models propose that the continuum arises from the dense-gas envelope, where hard ionizing radiation from the central engine is reprocessed into a stellar-like photosphere with an effective temperature of $\sim$5000 K. This implies that both the UV continuum and narrow-line emission are then powered by the host galaxy rather than an exposed central engine. To test whether this is consistent with the observed narrow-line ratios, we analyze multiple line diagnostics for a sample of $\sim$20 LRDs with high signal-to-noise NIRSpec grating spectra. We find that at least 40\% of the LRDs have line ratios pointing toward high ionization parameter and electron temperature, with a further 15\% also falling in the AGN regime for the O\textsc{i}/H$α$ diagnostic, indicative of harder ionizing radiation. These line ratios are incompatible with stellar photoionization from a star-forming host alone. This suggests lower density channels within the gas envelope through which high energy photons can escape and excite the surrounding narrow-line emitting gas. At the same time, most LRDs lack strong high-ionization line emission, with He\,\textsc{ii}/H$β$ $\lesssim0.1$, consistent with an ionizing spectrum softer than that of a standard AGN. Together, these results disfavour a uniform gas envelope with a covering fraction of unity, and instead point to a more complex geometry that gives rise to anisotropic ionizing radiation.

Short digest

This paper tests whether the narrow-line emission in Little Red Dots can be explained purely by host-galaxy star formation if their central engine is buried inside a nearly unity-covering, optically thick gas envelope. Using high-S/N JWST/NIRSpec grating spectra for a sample of about 20 LRDs and multiple narrow-line diagnostics, the authors find that at least 40% show line ratios requiring high ionization parameter and electron temperature, with a further 15% also entering the AGN-like O I/Hα regime. At the same time, most objects still lack strong high-ionization emission, with He II/Hβ remaining low, so the ionizing spectrum is softer than a standard unobscured AGN. The combination argues against a uniform fully closed envelope and instead favors a clumpy or anisotropic gas geometry with lower-density escape channels that let hard photons reach the narrow-line gas.

Key figures to inspect

• Figure 2. This is the core evidence figure because it places the LRD sample on the narrow-line diagnostic diagrams used to separate stellar photoionization from harder ionizing sources. Readers can see which objects lie outside the star-forming locus and how the measurements compare to stellar photoionization grids in ionization parameter and metallicity, making the paper’s main claim visually explicit.
• Figure 4. This figure is the best synthesis of how the different diagnostics agree or disagree across the sample. The UpSet layout makes the lower-limit fractions easy to read, including the subset with high-ionization and high-electron-temperature signatures and the smaller subset where all three diagnostics collectively point to AGN or shocks, which is central to the paper’s population-level conclusion.
• Figure 5. This schematic is the physical interpretation figure that translates the line-ratio results into a gas-geometry picture. It is the most direct visualization of the paper’s bottom line that LRDs likely have high-column regions producing the red continuum and Balmer-break-like features, plus lower-density channels that allow harder radiation to escape and power AGN-like narrow lines.
• Figure 3. This figure is useful for showing object-by-object consistency rather than only aggregate fractions. The top and bottom panels quantify how many independent diagnostics tag each LRD as AGN-like and what fraction of available diagnostics do so, helping readers judge how robust the classifications are across an inhomogeneous observational sample.