Too many or too massive? Investigating the high-$z$ demography of active SMBHs from JWST

TL;DR

This study confronts JWST-era measurements of high-redshift AGN demography by propagating the galaxy SMF into the BHMF via multiple $M_{ m BH}-M_{ m bm}$ relations, then into AGN LFs through an EDdington ratio distribution and a duty-cycle parameter. By integrating the continuity equation and applying abundance matching, the authors assess how initial SMBH–galaxy scaling and accretion histories reproduce the observed bolometric and UV AGN LFs and the SMBH mass density from $z\sim5.5$ to $z\sim0$. The results favour a scenario where SMBHs at $z\sim5$ lie modestly above local AGN scaling relations with elevated but plausible duty cycles, though a high-normalisation relation (e.g., $M_{ m BH}-M_{ m bm}$ significantly above local) can also fit data if duty cycles are low; the two pathways yield different evolutionary footprints for the BHMF and the $M_{ m BH}-M_{m}$ relation. Overall, the work highlights degeneracies between scaling relations and accretion histories and emphasizes that joint constraints from clustering, SMBH density, and multiwavelength LFs are needed to robustly map SMBH growth in the early universe.

Abstract

Recent JWST observations have unveiled a numerous population of low-luminosity active galactic nuclei (AGN) at $4< z<10$, with space densities roughly an order of magnitude above pre-JWST estimates, and many of these AGN have masses orders of magnitude above the local black hole mass-stellar mass ($M_{\rm BH}-M_{*}$) scaling relations. We investigate the consistency of these observations within a data-driven framework that links the galaxy stellar mass function to the supermassive black hole (SMBH) mass function and AGN luminosity functions using different $M_{\rm BH}-M_{*}$ relations and the observed Eddington-ratio distribution. By comparing our predictions against observed AGN luminosity functions at $z\sim 5.5$ we find that observations can be reproduced either by highly-elevated $M_{\rm BH}-M_{*}$ relations paired with low duty cycles, or moderate relations with higher duty cycles. Through the Soltan argument, we find that $M_{\rm BH}-M_{*}$ relations that are modestly above the local relation for AGN produce consistency between multiple tracers of the SMBH demography at $z\sim 5.5$, while more extreme normalisations would require a weakly-evolving luminosity function at $z> 5.5$. Continuity-equation modelling shows that initially high $M_{\rm BH}-M_{*}$ relations predict a strong two-phase evolutionary scenario and very steep low-mass SMBH mass functions in tension with several current estimates, while more moderate relations generate local SMBH mass functions in better agreement with present determinations and near-constant scaling relations. Our results favour a scenario where SMBHs at $z \sim 5$ on average lie modestly above local AGN scaling relations, with elevated but physically plausible duty cycles. Future wide-field clustering and demographic studies will help break the remaining degeneracies between SMBH scaling relations and AGN duty cycles at early cosmic times.

Figures (20)

• Figure 1: Top: A comparison of our reference LF which is a modified version of the Shen_2020 global fit A to the high-$z$ data it was tuned to 2023ApJ...959...39Hmaiolino2024jadesdiversepopulationinfant2025IgnasarXiv250403551J2011GlikmanApJ...728L..26G2022GrazianQUBRICsApJ...924...62G2023GrazianRubiconApJ...955...60G2021WolfAA...647A...5W2023BarlowHallMNRAS.519.6055B2025BarlowHallarXiv250616145B and the fiducial Shen_2020 global fit A. We also include the intermediate UV LF of 2024GrazianApJ...974...84G. Here, the X-ray and UV-based estimated have been converted to the bolometric plane following Shen_2020. A comparison between our reference LF and several theoretical models can be found in Fig. \ref{['fig:SAMcomp_MFLF']}. Bottom: A comparison of the Kokorev2024_PhotoConcensusOfLRDs LF (red square points) to our best fit Schechter function (red dotted line) and its $1\sigma$ uncertainty region (red shaded area), and the left-pointing arrows denote the $\sim1~{\rm dex}$ correction suggested by 2025GreenearXiv250905434G. We have also included the akins2024LRDs LRD LF (dark red circular points), the LFs of 2025GreenearXiv250905434G (orange hexagonal points) and Mathee2024_LRDs (dark orange diamond points), and the Shen_2020 quasar LF (grey shaded region).
• Figure 2: A pictorial representation of our methodology. By moving from left to right, we convert galaxy number densities to AGN demography, which are both independently calibrated quantities. The input ingredients (red boxes) included in each step of the chain are also informed by observations. The match with the observed AGN luminosity function and active SMBH mass function can reveal potential inconsistencies and/or biases in any of the input quantities. The comparison with the SMBH mass densities in the bottom row, can instead constrain the energy efficiencies of SMBHs (Soltan argument, extended to $z\sim10$) and provide an overall and self-consistent view of the accretion histories of SMBHs from very early epochs.
• Figure 3: A comparison of the Reines_2015, Pacucci2023_LRDs, and 2025LiTipOf$M_{\rm BH}-M_{\star}$ relations with the local sample of AGN from Reines_2015 (black points) and the faint AGN samples of maiolino2024jadesdiversepopulationinfant, 2023ApJ...959...39H, and 2023Ubler. The grey ellipses are the $1\sigma$, $2\sigma$, and $3\sigma$ confidence ellipses for the Reines_2015 sample.
• Figure 4: A plot of the $L_{\rm bol}-L_{\rm bol}/L_{\rm UV}$ plane comparing the individual BL AGN (points) to the power law fit of this population (black solid line), the relation obtained from abundance matching (AM, brown dashed line), the Shen_2020 UV-bolometric correction (dark red-dot dashed line). We also include a sample of high-$z$ quasars (black points) from 2019OnoueApJ...880...77O, 2019MatsuokaApJ...872L...2M, and 2010WillottAJ....140..546W. The quasar's $L_{\rm bol}$ is computed from the monochromatic luminosity $L_{3000}$ adopting the bolometric correction of 2006RichardsApJS..166..470R, which matches 2011ShenApJS..194...45S. The filled BL AGN data points 2023ApJ...959...39Hmaiolino2024jadesdiversepopulationinfant2025IgnasarXiv250403551J use the observed $L_{\rm UV}$, whereas the open symbols use the intrinsic $L_{\rm UV}$. The LRD data points Mathee2024_LRDsKokorev2024_PhotoConcensusOfLRDs2024arXiv240403576K use the fiducial $L_{\rm bol}$. For the clarity of the plot, we have not plotted the uncertainties in the data points, and instead display the median uncertainty as the grey error bar on the centre left. The uncertainties for the individual objects were included when fitting the $L_{\rm bol}-M_{\rm UV}$ relation.
• Figure 5: A plot of the BHMF at $z=5.5$ obtained from the SMF assuming the three $M_{\rm BH}-M_{\star}$ relations: Reines_2015 (solid dark blue line), Pacucci2023_LRDs (dashed light blue line), 2025LiTipOf (dot-dashed mint green solid line). These total BHMFs are compared to the observed active BHMFs of BL AGN. We have also included the BHMFs scaled by the AGN fraction that best aligns them with the active BHMF of taylor2024broadlineagn35z6black, these AGN fractions are $f_{\rm AGN}\sim0.08,~0.5$ for the Pacucci2023_LRDs, and 2025LiTipOf relations, respectively, and an AGN fraction in excess of unity is required for Reines_2015. These scaled BHMFs are denoted by the lower-opacity, thinner lines of the same color and linestyle. The high-$z$ active BHMFs included are those of taylor2024broadlineagn35z6black, 2025GerisarXiv250622147G, 2025FeiarXiv250920452F, and 2024HeApJ...962..152H. A comparison between the BHMFs obtained from the SMF assuming the three $M_{\rm BH}-M_{\star}$ relations and several theoretical models can be found in Fig. \ref{['fig:SAMcomp_MFLF']}.