BlackTHUNDER -- A non-stellar Balmer break in a black hole-dominated little red dot at $z=7.04$

TL;DR

This study uses new high-resolution JWST/NIRSpec-IFU data plus archival imaging and spectra to investigate Abell 2744-QSO1, a z=7.04 LRD with a Balmer-break that challenges stellar interpretations. By modeling the optical continuum as AGN light attenuated by dense, dust-free BLR gas and a separate dust screen, the authors reproduce a smooth Balmer break and a tentative Hβ absorption, arguing for a non-stellar origin of the break. They derive a black hole mass of about 4×10^7 M⊙ with sub-Eddington accretion, and set a tight dynamical-mass upper limit for the host (~4×10^8 M⊙), implying the BH is overmassive relative to both stellar and dynamical host masses. The work further discusses variability and reverberation-mapping potential, the broader applicability to other LRDs, and the implications for X-ray weakness and early BH growth, offering a framework that mitigates extreme stellar-density requirements for Balmer-break interpretations at high redshift.

Abstract

Recent observations from JWST have revealed an abundant population of active galactic nuclei (AGN) and so-called ``Little Red Dots'' (LRDs) at $2\lesssim z \lesssim 11$, many of which are characterized by V-shaped UV-to-optical continua with turnovers around the Balmer limit. The physical nature of these LRDs is unclear, and it remains debated whether the peculiar spectral shape originates from AGN, compact galaxies, or both. We present the analysis of new NIRSpec-IFU data from the BlackTHUNDER JWST Large Programme and archival NIRSpec-MSA data of a lensed LRD at $z=7.04$. The spectra confirm the presence of a smooth Balmer break and a broad H$β$ tracing the Broad Line Region (BLR) of an AGN. The small velocity dispersion of the H$β$ narrow component indicates a small dynamical mass of the host galaxy of $M_{\rm dyn}<4 \times 10^8~M_{\odot}$, which implies that the stellar population cannot contribute more than 10% to the optical continuum. We show that the Balmer break can be well described by an AGN continuum absorbed by very dense ($n_{\rm H}\sim 10^{10}~{\rm cm^{-3}}$) and nearly dust-free gas along our line-of-sight (possibly gas in the BLR or its surrounding). The same gas is expected to produce H$β$ absorption, at a level consistent with a tentative detection ($3σ$) in the high-resolution spectrum. Such a non-stellar origin of the Balmer break may apply to other LRDs, and would alleviate the issue of extremely high stellar mass surface densities inferred in the case of a stellar interpretation of the Balmer break. We note that this is a rare case of a black hole that is overmassive relative to both the host galaxy stellar and dynamical masses. We finally report indications of variability and the first attempt of AGN reverberation mapping at such an early epoch.

Figures (24)

• Figure 1: Observations of the triply imaged Abell 2744-QSO1. Top: Color-composite NIRCam images as well as NIRSpec/IFU maps. The NIRCam images of three lensed images from the UNCOVER DR4 uncover_dr4 are shown, where the field of view of the IFU observation for image A is indicated by the dotted white rectangular aperture. For IFU maps, fluxes of the narrow and the whole H$\beta$ are shown and the extraction aperture for the PRISM and grating spectra is indicated by the dashed white circle. Middle: IFU and MSA PRISM spectra combining different lensed images normalized to the flux density at $\lambda = 4260$ Å in the rest frame. The spectra are manually shifted in the $y$ axis for presentation purposes. From top to bottom, the spectra correspond to the central region of image A extracted from a circular aperture with a diameter of $0.\!\!^{\prime\prime}25$ using the BlackTHUNDER IFU observations with the GTO reduction, the A and C images combined from the MSA observations with the GTO reduction (image B is removed due to the background subtraction issue described in Section \ref{['sec:data']}), and the A, B, and C images combined from the MSA observations with furtak2024's reduction, respectively. The shaded regions correspond to the $1\sigma$ pipeline uncertainties. Bottom: The high-resolution (R2700) BlackTHUNDER IFU spectrum of image A extracted from a circular aperture of $0\farcs25$ zoomed around the location of H$\beta$ and [O iii]$\lambda \lambda 4960,5008$. The dashed lines are the best-fit emission line and continuum models, which show a tentative absorption in H$\beta$ at a significance of $3\sigma$. The two bottom panels show the residuals normalized by uncertainties ($\chi={\rm residual}/\sigma$) of fits with and without the H$\beta$ absorption, respectively. The fit with the H$\beta$ absorption produces improved $\chi$ residuals near the centroid of the broad H$\beta$ ($\sim 3.91~{\rm \mu m}$).
• Figure 2: Comparison of stellar and dynamical masses for AGN at high-$z$ discovered by JWST and observed with high spectral resolution. The shaded region is the "unphysical" area where the stellar mass is larger than the dynamical mass. Red small symbols are objects in the JADES sample presented in maiolino2023b (updated with new estimated stellar masses in juodzbalis_jadesbh_2025). The upper golden symbol shows the dynamical mass estimated by the BlackTHUNDER high resolution spectrum of Abell2744-QSO1, where the stellar mass is the one inferred by mayilun_lrd_2024 assuming that the optical continuum and Balmer break are entirely dominated by stellar light; clearly this scenario leads to an unphysically high stellar mass, about an order of magnitude larger than the dynamical mass. The maximum stellar mass allowed by the dynamical mass is shown with the lower golden symbol.
• Figure 3: Scaling relations between black holes and their host galaxies, specifically BH mass versus galaxy stellar mass (left) and versus galaxy dynamical mass (right). The small blue symbols show local galaxies, and the straight lines and shaded regions illustrate the best fit local scaling relations (see text for details). The red points show AGN at 4$<$z$<$11 for which the black hole mass and host galaxy stellar/dynamical mass has been measured with JWST data, as reported by Bogdan23Carnall+2023Goulding23Ding23harikane2023kocevski2023Kokorev2023ubler2023amaiolino2023bStone_2024Yue_bhmass_2024wang_break_2024trefoloni_feii_2024juodzbalis_jadesbh_2025. The mean uncertainties of masses are indicated by the black cross on the top left location of each panel. Additionally, we plot high-$z$ sources from Izumi_2019Pensabene_2020Akins_ci_2025. The golden symbols show Abell2744-QSO1 for which we have taken the upper limit on the host galaxy dynamical mass as conservative upper limit on the stellar mass. The black hole in Abell2744-QSO1 is clearly overmassive both in terms of stellar and dynamical mass, when compared with the local relations.
• Figure 4: Example of an AGN accretion disk emission with $M_{\rm BH}=10^7~M_{\odot}$ and $\lambda _{\rm Edd}=0.1$ attenuated by a slab of gas with $n_{\rm H}=10^{10}~{\rm cm^{-3}}$, $N_{\rm H}=10^{23}~{\rm cm^{-2}}$, $U=10^{-1.5}$, and $v_{\rm turb}=120$$\rm km~s^{-1}$ (definitions of parameters given in Section \ref{['subsec:params_dependency']}). The simulation is performed with Cloudy. The dashed line is the intrinsic continuum normalized at $\lambda =4260$ Å. The dash-dotted red line is the dust-free attenuated continuum, which exhibits prominent absorption features including Balmer lines and the Balmer break. The solid blue line is the attenuated continuum further obscured by a dust screen with $A_{\rm V}=2$, which shows an uprising optical continuum in addition to the Balmer break resembling the rest-frame optical part of LRDs. The spike at the UV-optical interface is an artifact caused by the limited number of resolved energy levels of hydrogen.
• Figure 5: Dependencies of the strength of the Balmer break and the equivalent width of the H$\beta$ absorption on various model parameters for an AGN continuum attenuated by a slab of dust-free gas computed with Cloudy. From top left to bottom right, we vary the hydrogen density, $n_{\rm H}$, the hydrogen column density, $N_{\rm H}$, the ionization parameter, $U$, and the microturbulence velocity, $v_{\rm turb}$, respectively. The horizontal red dashed lines and shaded regions represent lower limits and their 68% confidence intervals constrained by the observed spectra of Abell 2744-QSO1. The vertical green dashed lines and shaded regions represent the upper limit on $v_{\rm turb}$ and its 68% confidence interval. For the strength of the Balmer break, the constraint is a lower limit since the intrinsic break is lifted by the extension of the UV continuum. Similarly, for EW(H$\beta$), the constraint is a lower limit since the optical tail the UV spectrum is not subtracted. The upper limit on $v_{\rm turb}$ is obtained from the measured width of the H$\beta$ absorption. The inclusion of the H$\beta$ absorption tightens the constraints on the microturbulence velocity, $v_{\rm turb}$, assuming it originates in the same absorber that produces the Balmer break.