The Small Sizes and High Implied Densities of `Little Red Dots' with Balmer Breaks Could Explain Their Broad Emission Lines Without an AGN

TL;DR

This study targets three Balmer-break galaxies at z∼6–8 that JWST images reveal as extraordinarily compact, with rest-frame UV light dominated by an older stellar population, yet exhibiting broad Hβ emission.By combining precise size measurements from F200W imaging with spectral decompositions that consider varying AGN contributions, the authors derive three mass scenarios and compute central densities, offering a test between AGN-dominated and purely stellar kinematic interpretations of the broad lines.The results show that no-AGN masses imply central densities and predicted dispersions beyond typical galaxies, while the observed Hβ widths can be explained by virial motions in the most massive (no-AGN) scenario, leaving open the possibility that some broad lines are AGN-driven.The work highlights extreme densities and compactness as a potential clue to rapid early assembly or alternative evolutionary paths, while outlining critical follow-up observations needed to discriminate between the competing scenarios and to understand the connection to present-day massive galaxies.

Abstract

Early JWST studies found an apparent population of massive, compact galaxies at redshifts $z\gtrsim7$. Recently three of these galaxies were shown to have prominent Balmer breaks, demonstrating that their light at $λ_{\rm rest} \sim 3500$ $Å$ is dominated by a stellar population that is relatively old ($\sim$200 Myr). All three also have broad H$β$ emission with $σ> 1000 \,\rm km s^{-1}$, a common feature of such `little red dots'. From Sérsic profile fits to the NIRCam images in F200W we find that the stellar light of galaxies is extremely compact: the galaxies have half-light radii of $r_{\rm e}\sim$ 100 pc, in the regime of ultra compact dwarfs in the nearby Universe. Their masses are uncertain, as they depend on the contribution of possible light from an AGN to the flux at $λ_{\rm rest}>5000$ $Å$. If the AGN contribution is low beyond the Balmer break region, the masses are $M_* \sim 10^{10}-10^{11}\mathrm{M}_{\odot}$, and the central densities are higher than those of any other known galaxy population by an order of magnitude. Interestingly, the implied velocity dispersions of $\sim$1500 kms$^{-1}$ are in very good agreement with the measured H$β$ line widths. We suggest that some of the broad lines in `little red dots' are not due to AGNs but simply reflect the kinematics of the galaxies, and speculate that the galaxies are observed in a short-lived phase where the central densities are much higher than at later times. We stress, however, that the canonical interpretation of AGNs causing the broad H$β$ lines also remains viable.

Figures (7)

• Figure 1: The spectra (grey) and best-fit model (darkblue) with maximal stellar mass of three Balmer break galaxies, obtained from Wang2024_balmer. We also show $1\hbox{$^{\prime\prime}$} \times 1\hbox{$^{\prime\prime}$}$ images in four different bands, F115W, F200W, F356W and F444W. The dotted lines indicate the location of the Balmer break (black), [OIII] line (red), and H$\beta$ (blue) line, which are relevant in this work. For information on other emission lines, see Wang2024_balmer.
• Figure 2: The top row shows images in the NIRCam F200W band of the three Balmer break galaxies. The second row shows the segementation maps, the third the best-fitting PSF convolved galfit models. The residuals, obtained by subtracting the best-fitting models from the images, are shown in the fourth row. The fifth row shows RGB images of the galaxies, with F115W as the blue band, F150W as the green band and F200W as the red band and the final row with F115W, F277W, F444W. All images are $1\hbox{$^{\prime\prime}$} \times 1\hbox{$^{\prime\prime}$}$.
• Figure 3: The size measured in F200W, plotted against the UV magnitude, measured from the spectrum at $\lambda_{\rm rest}$=150 nm. We also show size-luminosity derived for local spiral galaxies from DeJong2000 for which we use a magnitude correction $\rm M_{\rm UV}$-$\rm M_{\rm I}=1.0$, and LBGs at $z\sim8$Shibuya2015Bouwens2022Kawamata2018, and $z\sim9-12$Yang2022. In addition we show star-forming complexes Chen2023 with $z_{\rm phot}\sim6-8$. Finally, we show extremely compact sources, recently detected through lensing with JWST, reported in Furtak2024 ($R_{\rm e}<35$ pc, $z=7.0$), Topping2024 ($R_{\rm e}<22$ pc, $z=6.1$) and Williams2023_science ($R_{\rm e} = 16$ pc, $z=9.5$). Finally we show the upper limit of $\lesssim$100 pc reported for the brightest LRDs Akins2024 (see text).
• Figure 4: The surface density ($\Sigma_{\mathrm{eff}} = M_{*}(<R_{\mathrm{e}})/\pi R_{\mathrm{e}}^2$) plotted against stellar mass. We show a theoretical limit $\Sigma_{\mathrm{eff}} \leqq 3\times 10^5 \mathrm{M}_{\odot} \mathrm{pc}^{-2}$, determined by Grudic2019 as the dashed black line. For illustration, we show regions in $M_*-\Sigma_{\mathrm{eff}}$ space for multiple stellar systems estimated from Hopkins2010_density, the class of star clusters (SC), ultracompact dwarf galaxies (UCD) and globular clusters (GCs) (green), local elliptical galaxies (grey) and compact $z>2$ elliptical galaxies (yellow). We also show (see text) quiescent galaxies at $z\sim4-5$Setton2024Graaff2024Carnall2023_nature, compact starburst galaxies at $z=0.4-0.8$Stanic2021, Cosmic Gems star clusters at $z\sim10$Adamo2024 and the MW nuclear star cluster. The surface density measured for the three Balmer break galaxies are shown for the three sets of stellar masses $M_{*,\mathrm{min}}$, $M_{*,\mathrm{med}}$ and $M_{*,\mathrm{max}}$ in orange, red and purple, respectively.
• Figure 5: The median stellar mass profiles for the three sets of stellar masses (see Table \ref{['tab:sizes']}), $M_{*,\mathrm{min}}$, $M_{*,\mathrm{med}}$ and $M_{*,\mathrm{max}}$ in orange, red and purple, respectively. We also show the stellar mass profiles of massive quiescent galaxies at different cosmic times: the mean stellar mass profile of three compact massive quiescent galaxies at $z\sim4-5$Carnall2023_natureSetton2024Graaff2024, $z\sim2.3$ compact elliptical galaxies Bezanson2009, and massive elliptical galaxies at $z=0$Tal2009.