Back to Normal Again: Possible Destinies of JWST overmassive SMBHs and "Little Red Dots" in the View of Shin-Uchuu Simulation

Abstract

The James Webb Space Telescope (JWST) has enabled the discovery of hundreds of supermassive black holes (SMBHs) at redshifts $z\gtrsim 4-7$. A non-negligible fraction of these SMBHs are hosted in galaxies with BH-to-galaxy mass ratios ($M_{\rm BH}/M_\star$) being excessively larger than that for local SMBHs by $\sim 1-2$ dex. The origin of these ``overmassive'' BHs remains elusive, demanding either a heavy seed formation scenario or rapid growth of seed BHs. Their deviation from local scaling relations challenges our understanding of how SMBHs and their host galaxies coevolve across cosmic time. In this paper, we apply phenomenological modelings for BHs and galaxies to dark matter halo merger histories from N-body simulations to investigate the subsequent evolution of JWST-discovered ``overmassive'' SMBHs. We find that early evolution of ``overmassive'' SMBHs is dominated by stunted accretion leading to gradual decreases in $M_{\rm BH}/M_\star$ ratios. In contrast, less massive SMBHs experience super-Eddington accretion during their early evolution, resulting in a slow increase of mass ratios toward $M_{\rm BH}/M_\star \sim 0.01$. Convergence occurs at $M_{\rm BH}\sim 10^8~M_\odot$ with $M_{\rm BH}/M_\star \sim 0.01$. At lower redshift, nearly all SMBHs evolve onto local relations, as expected given that our models adopt empirical relations derived from low-redshift observations. This suggests that the global feedback mechanisms regulating the coevolution of $M_{\rm BH}/M_\star$ ratios are implicitly encoded in local relations in terms of star-formation rate distribution, black hole accretion rate distribution and their active (quiescent) fractions.

Figures (6)

• Figure 1: This figure illustrates functional forms for quiescent fractions of galaxies (dashed curves) and active fraction of BH activities (with Eddington ratio $\lambda > 0.01$, solid curves). For quiescent fraction, observational constraints inferred from different methods are overlaid as gray dashed curves for comparisons Leja+2022. For BH active fraction, models from observational indications at different redshift bins are overlaid in dash-dotted curves Schulze+2015. For simplicity, a two-slope functional form with moderate redshift evolution for active fractions is adopted (solid curves). The absolute values of active fraction at high mass end is small, the exact function forms won't modify our results significantly.
• Figure 2: The mass evolution for selected halos and corresponding galaxy model from UniverseMachine, with stellar mass of SMBH and LRD sample overlaid in filled orange circles. DM halo merger trees are selected based on stellar mass criterion with a variation $\pm 10\%$. For each object, $50$ merger trees are selected. For some high-stellar-mass samples with fewer merger trees, the selection criterion is eased to variation of $1$ dex to reach $50$ merger trees. Note for object J0100+2802, only $9$ merger trees are extracted.
• Figure 3: The mass evolution for the selected four sources with different levels of "Overmassiveness" (from left to tight): GN-z11, CEERS-20496, UNZ1 and Abell-2744-QSO1. The red, blue and orange colors represent mass evolution for DM halos, for galaxies and for BHs. For BH mass evolution, accretion epochs with super-Eddington rates and at "Quasar" mode are highlighted in red and orange stars, while rest epoche are at "Radio" mode. For better visualization, only $15$ trees out of $50$ are shown in these plots.
• Figure 4: Upper panel: The SFR distribution as a function of stellar mass for our galaxy model (blue dots and blue contours) and for UM galaxy model (gray dots and gray contours). For clarity, only galaxy models for $4$ selected samples are shown in both plots. The black curve shows the fitted relation for $z\sim 0$ galaxy samples from SDSS in Peng+2010. Lower panel: The stellar-to-halo-mass (SMHM) ratio for $4$ selected samples, for our model (blue curves) and for galaxy formation model from UniverseMachine (gray curves). The black curve represents the inferred median stellar-to-halo-mass ratio at $z=0$ in UM Behroozi+2019.
• Figure 5: Top panels: The evolution of BH-to-galaxy-mass ratios for the four selected sources (left panel) and for the whole sample (right panel). In top-left panel, initial conditions for the four sources are highlighted in black stars, while their evolutions are shown in solid curves with redshift color-coded. At $z=0$, their evolution destinations are highlighted in blue stars. As comparisons, other high-$z$ JWST SMBHs and LRDs Ding+2023Harikane+2023Kocevski+2023Larson+2023Ubler+2023Maiolino+2024Maiolino+2024bJuodzbalis+2024Bogdan+2024Ubler+2024Stone+2024Yue+2024Greene+2024Kocevski+2025 are shown in color-coded stars and local relation from Kormendy+2013 and characteristic ratios ($M_{\rm BH}/M_\star=10^{-3},~10^{-2},~10^{-1}$) are overlain in black curve and in gray curves. Lower panel: The MM ratio evolution as a function of redshifts for the whole sample and for other high-$z$ SMBHs and LRDs. The initial conditions for $26$ selected sources are highlighted in red stars. The solid blue curves are individual evolution of MM ratios for all sources, while the solid red curves represent the averaged MM ratio evolution. The dashed, dashed–dotted, and dotted horizontal lines are BH-to-galaxy-mass ratios for late-type, early-type, and all galaxies at $M_\star=3\times 10^{10}~M_\odot$Greene+2020, respectively. The black solid curve shows the upper limits for MM ratio in BH growth model from Hu+2025a, while orange stars are for high-$z$ JWST SMBHs and LRDs.