Dynamical Modeling of the Broad-Line Region with High-Mass Active Galactic Nuclei and Constraints on the Virial Factor

arxiv: 2604.24721 · v1 · submitted 2026-04-27 · 🌌 astro-ph.GA

Dynamical Modeling of the Broad-Line Region with High-Mass Active Galactic Nuclei and Constraints on the Virial Factor

Shu Wang , Jong-Hak Woo , Lizvette Villafa\~na , Tommaso Treu , Elena Gallo This is my paper

Pith reviewed 2026-05-08 02:29 UTC · model grok-4.3

classification 🌌 astro-ph.GA

keywords broad-line regionactive galactic nucleiblack hole massvirial factorreverberation mappingdynamical modelingM_BH-sigma relation

0 comments p. Extension

The pith

Dynamical modeling of broad-line regions in high-mass AGNs calibrates the virial factor to log10(f) = 0.69 ± 0.21.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper performs full dynamical modeling on the broad-line regions of eight high-mass active galactic nuclei to recover their black hole masses directly. It compares those masses against simpler virial products calculated from reverberation time lags and emission-line widths, thereby determining the scaling factor f required for mass estimates in other AGNs. When the new objects are combined with thirty earlier ones for a total sample of thirty-eight, the authors obtain average recommended values of f that match those found by aligning reverberation-mapped AGNs with the black-hole-mass versus stellar-velocity-dispersion relation measured in quiescent galaxies. The work extends reliable dynamical constraints into the mass range 10^8 to 10^8.5 solar masses where the virial factor had remained poorly determined.

Core claim

The Hβ-emitting broad-line regions are best described as thick disks viewed at intermediate inclinations, with emission preferentially from the far side and kinematics that include rotational, inflowing, and outflowing components. Dynamical modeling yields black-hole masses that, when compared to reverberation-based virial products, give individual virial factors. For the combined sample of thirty-eight objects the predictive values are log10(f_pred) = 0.69 ± 0.21 using rms line dispersion and −0.08 ± 0.23 using mean FWHM. These values align with those obtained by matching the AGN sample to the M_BH–σ* relation of quiescent galaxies, and the intrinsic scatter around the mean f is ~0.2 dex.

What carries the argument

BLR dynamical modeling that solves simultaneously for thick-disk geometry, viewing inclination, and mixed kinematic components (rotation plus radial flows) to recover the true black-hole mass, which then calibrates the virial factor f = M_BH / (R ΔV² / G).

If this is right

The ~0.2 dex intrinsic dispersion in the new f values permits more precise single-epoch black-hole mass estimates than those based solely on the M_BH–σ* relation.
The sample now covers the previously poorly constrained interval 10^8 to 10^8.5 solar masses, allowing the AGN M_BH–σ* relation to be tested at the high-mass end.
Consistency between the dynamical f and the alignment with quiescent-galaxy M_BH–σ* supports the assumption that local active and inactive galaxies obey the same scaling relation.
High-mass AGNs exhibit thick-disk BLR geometry with far-side emission preference and mixed rotational plus radial motions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Adopting the new f values for large spectroscopic surveys would reduce systematic uncertainty in black-hole demographic studies at higher redshifts where full dynamical modeling is infeasible.
If the virial factor remains mass-independent, the same scaling can be applied uniformly across AGN samples to trace black-hole growth over cosmic time.
Repeating the dynamical analysis on still more massive or differently selected AGNs would test whether f depends on luminosity, Eddington ratio, or host-galaxy properties.

Load-bearing premise

The thick-disk geometry and chosen kinematic components assumed in the dynamical model accurately recover the true broad-line region structure and black-hole mass without large systematic biases from data quality or model parameterization.

What would settle it

Independent black-hole mass measurements (for example from stellar dynamics or maser kinematics) that systematically disagree with the dynamical modeling masses for these same eight AGNs would invalidate the derived average virial factor.

Figures

Figures reproduced from arXiv: 2604.24721 by Elena Gallo, Jong-Hak Woo, Lizvette Villafa\~na, Shu Wang, Tommaso Treu.

**Figure 1.** Figure 1: Spectral decomposition of mean spectra for the eight AGNs modeled in this work. The continuum of each quasar is decomposed into a power law component (green), an Fe pseudocontinuum (dark green) based on the template provided by Boroson & Green (1992), and a host galaxy component (brown). The emission-line components consist of Hβ (magenta), narrow (blue) and wing (gray) [O iii], and He ii (cyan) emission l… view at source ↗

**Figure 2.** Figure 2: Illustration of the BLR model used in CARAMEL. Left panel: projection onto the x–z. The dashed arrow indicates the direction of the observer’s line of sight. The BLR is viewed at an inclination angle of θi = 30◦ and has an opening angle of θo = 30◦ , defined with respect to the BLR disk midplane. The symbol size represents the relative line emissivity, demonstrating a case where the emission is stronger on… view at source ↗

**Figure 3.** Figure 3: The CARAMEL model fits to the Hβ emission-line profile, integrated Hβ light curves, and AGN continuum light curves. From top to bottom: Panels 1 and 2 show the observed and modeled Hβ line profiles, ordered by observation epoch. Panel 3 presents the normalized residuals (χ; see the main text for detailed description), while Panel 4 displays the observed Hβ profile for a randomly selected epoch alongside th… view at source ↗

**Figure 3.** Figure 3: Continued. The results for PG 1121+422, J1217+333, PG1427+480, and J1540+355 view at source ↗

**Figure 4.** Figure 4: Geometric representation of BLR emission for the eight AGNs, based on median parameter estimates from CARAMEL . The left panel depicts an edge-on perspective for each source, while the right panel shows a face-on view. Each point in the plot represents a point-like gas cloud, with colors indicating their dynamics: red for inflow and blue for outflow. clination angle of θi = 21.7 +9.8 −8.8 degrees relative … view at source ↗

**Figure 5.** Figure 5: Transfer function produced using median model parameter estimates. The right-hand panel shows the velocityintegrated transfer function and the bottom panel shows the average time lag for each velocity pixel. The results for J0140+234, PG0947+396, J1026+523, and J1120+423 are shown view at source ↗

**Figure 5.** Figure 5: Continued. The results for PG 1121+422, J1217+333, PG1427+480, and J1540+355 view at source ↗

**Figure 6.** Figure 6: The predictive distribution of virial factors (red dashed) along with the posterior distribution of the mean virial factors (blue solid). The gray histograms represent the distribution of virial factors calculated from individual sources. The results presented here are based on the 38 AGNs from the combined sample, and the four panels represent the results for different types of line widths. These dynamica… view at source ↗

**Figure 7.** Figure 7: Comparison between black hole masses derived from CARAMEL modeling and traditional RM-based estimates. Red and blue symbols denote the SAMP and literature samples, respectively. The RM-based masses are calculated by applying the updated virial factor from this work (log10(f)pred = 0.69 ± 0.21) to the virial products. Two sets of error bars are shown: solid bars represent statistical uncertainties, whi… view at source ↗

read the original abstract

We present the results of broad-line region (BLR) dynamical modeling for eight high-mass active galactic nuclei (AGNs) from the Seoul National University AGN Monitoring Project, by constraining BLR geometry and kinematics as well as black hole (BH) mass ($M_{\rm BH}$). We find that the H$\beta$-emitting BLRs are best described as thick disks viewed at intermediate inclinations, with emission preferentially originating from the far side of the BLR. BLR kinematics show a combination of rotational, inflowing and outflowing components. By comparing the $M_{\rm BH}$ from dynamical modeling with the virial products based on reverberation lags and line widths, we determine the virial factor $f$ for individual AGNs. Combining our sample with those $M_{\rm BH}$ consistently determined from BLR dynamical modeling, yielding a total of 38 objects, we derive a virial factor for future $M_{\rm BH}$ estimation of log$_{10}({f})_{\rm pred}=0.69\pm0.21$ based on $\sigma_{\rm line,rms}$ and $-0.08\pm0.23$ based on FWHM$_{\rm mean}$. The derived virial factor is consistent with that inferred by aligning the reverberation-mapped AGNs with quiescent galaxies in the $M_{\rm BH}$-$\sigma_{\ast}$relation, supporting the assumption that local active and inactive galaxies follow the same $M_{\rm BH}$-$\sigma_{\ast}$ relation. Our updated $f$ values exhibit an intrinsic dispersion of $\sim0.2$ dex, which allows for a more precise $M_{\rm BH}$ estimates than those based on the $M_{\rm BH}$-$\sigma_{\ast}$ relation. Our sample extends the dynamical modeling-based reverberation sample to $M_{\rm BH}$ $\sim$ [$10^8$, $10^{8.5}$] $M_{\odot}$ range, where the virial factor from the the AGN $M_{\rm BH}$-$\sigma_{\ast}$ relation remains poorly constrained, underscoring the unique value of dynamical modeling analysis in constraining the $M_{\rm BH}$ of the most massive BHs.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This paper adds eight new high-mass AGNs to the dynamical BLR modeling sample and derives an updated average virial factor of log10(f) = 0.69 ± 0.21, extending coverage where the M_BH–σ* relation is weakest.

read the letter

The main point is that they model eight new high-mass objects from the SNU AGN Monitoring Project, combine them with prior dynamical results for a total of 38, and report a calibrated f that is tighter than the scatter in the M_BH–σ* relation. The models favor thick disks at intermediate inclinations with far-side emission and a mix of rotation, inflow, and outflow. They also show the new f aligns with the value obtained by matching reverberation-mapped AGNs to the quiescent galaxy M_BH–σ* relation. This is useful because it targets the 10^8–10^8.5 M⊙ range where other calibrations are sparse and gives a practical number for future mass estimates with ~0.2 dex intrinsic dispersion. The work is straightforward in its approach: direct comparison of dynamical M_BH to virial products from the same objects, without circularity in the primary derivation. The consistency check with the M_BH–σ* alignment is presented as supporting evidence rather than an input. On the downside, the modeling choices—thick-disk geometry, kinematic decomposition, and fixed far-side emission—could introduce net bias in the recovered M_BH, particularly for these higher-mass targets where data quality and degeneracies are less well tested. The abstract gives no fit statistics, uncertainties on individual geometries, or mock-data recovery tests, so it is hard to judge how robust the individual M_BH values are before they feed into the average f. If those checks are solid in the full text, the result holds; if not, the quoted uncertainty on f may be optimistic. This is worth a reading group slot for anyone working on AGN black-hole masses or reverberation mapping. It deserves peer review because the new data points and the expanded sample are concrete additions to the literature, even if referees will want to see the modeling diagnostics in detail.

Referee Report

2 major / 1 minor

Summary. This paper performs dynamical modeling of the broad-line region (BLR) in eight high-mass active galactic nuclei (AGNs) from the Seoul National University AGN Monitoring Project. It determines that the Hβ BLRs are best fit by thick disk geometries at intermediate inclinations with emission from the far side and kinematics including rotation, inflow, and outflow. Individual virial factors f are calculated by comparing dynamical black hole masses to virial products from reverberation mapping. The sample is combined with previous dynamical modeling results to reach 38 objects total, from which average virial factors are derived: log10(f)_pred = 0.69 ± 0.21 using σ_line,rms and -0.08 ± 0.23 using FWHM_mean. These are shown to be consistent with the value from aligning reverberation-mapped AGNs with the M_BH–σ* relation of quiescent galaxies.

Significance. Should the modeling assumptions hold without net bias, the work significantly extends the dynamically modeled reverberation sample into the poorly constrained high-mass regime (10^8 to 10^8.5 M_⊙). The resulting f calibration with ~0.2 dex intrinsic scatter offers a path to more precise single-epoch black hole mass estimates than those relying solely on the M_BH–σ* relation. The consistency with the M_BH–σ* alignment provides an important cross-check, and the explicit reporting of the combined sample strengthens the statistical basis for the recommended f values.

major comments (2)

The abstract reports best-fit geometries, kinematics, and derived f values, but provides no quantitative details on model fits, uncertainties, goodness-of-fit metrics, or tests against simulated data, leaving open whether the central claims on geometry and f are fully supported by the observations.
The headline result log10(f)pred=0.69±0.21 based on σline,rms requires that the BLR dynamical models recover true M_BH without net bias. The models adopt a thick-disk geometry, decompose velocity into rotation + inflow + outflow, fix emission to the far side, and fit intermediate inclinations. For the high-mass regime where data quality and model degeneracies are least tested, even modest systematic offsets in individual M_BH would shift the weighted mean f and its quoted uncertainty.

minor comments (1)

The notation 'log10(f)pred' and line-width symbols (e.g., σline,rms) should be explicitly defined at first use and used consistently throughout the manuscript.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript. Their comments highlight important aspects of clarity and robustness that we will address in the revision. Below we respond point by point to the major comments.

read point-by-point responses

Referee: The abstract reports best-fit geometries, kinematics, and derived f values, but provides no quantitative details on model fits, uncertainties, goodness-of-fit metrics, or tests against simulated data, leaving open whether the central claims on geometry and f are fully supported by the observations.

Authors: We agree that the abstract would be strengthened by the inclusion of quantitative summary information. In the revised manuscript we will expand the abstract to report key goodness-of-fit metrics (e.g., reduced chi-squared values for the best-fit models) and representative uncertainties on the derived parameters. The full quantitative details on model fits, uncertainties, and comparisons to simulated data are already presented in Sections 3 and 4 of the main text; we will add a brief cross-reference in the abstract to direct readers to these sections. This change will make the abstract more self-contained while preserving its length constraints. revision: yes
Referee: The headline result log10(f)pred=0.69±0.21 based on σline,rms requires that the BLR dynamical models recover true M_BH without net bias. The models adopt a thick-disk geometry, decompose velocity into rotation + inflow + outflow, fix emission to the far side, and fit intermediate inclinations. For the high-mass regime where data quality and model degeneracies are least tested, even modest systematic offsets in individual M_BH would shift the weighted mean f and its quoted uncertainty.

Authors: We acknowledge the referee's concern regarding possible net bias in the high-mass regime. Our modeling framework follows the same methodology validated in prior dynamical modeling studies across a range of black-hole masses, and the resulting average f is independently consistent with the value obtained by aligning the reverberation-mapped sample with the M_BH–σ* relation of quiescent galaxies. This cross-check provides empirical support that any residual bias is not large enough to dominate the result. Nevertheless, we agree that explicit discussion of data-quality limitations and model degeneracies specific to the 10^8–10^8.5 M_⊙ range is warranted. In the revision we will add a new subsection addressing potential systematic offsets, including a quantitative estimate of how modest biases would propagate into the weighted mean and its uncertainty. The reported intrinsic scatter of ~0.2 dex already reflects the observed dispersion; we will clarify this point as well. revision: partial

Circularity Check

0 steps flagged

No significant circularity: virial factor derived by direct ratio of independent dynamical masses to virial products

full rationale

The central result computes f by comparing M_BH values obtained from BLR dynamical modeling (thick-disk geometry, rotation+inflow+outflow kinematics, far-side emission) against the virial products R ΔV²/G measured from the same reverberation data. This ratio is not self-definitional or fitted-by-construction; the dynamical models solve for mass, geometry, and kinematics from the line-profile and lag data without presupposing the value of f. The sample is then combined with 30 prior objects that used the identical modeling approach, yielding a weighted mean f. The noted consistency with the M_BH–σ* alignment is presented only as a supporting check, not as an input or constraint on the derived f. No load-bearing self-citation, ansatz smuggling, or renaming of a known result occurs in the derivation chain. The paper is therefore self-contained against external benchmarks for the purpose of circularity analysis.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claims rest on the validity of the BLR dynamical modeling assumptions and the representativeness of the combined sample; limited information is available from the abstract alone.

free parameters (1)

average virial factor f = log10(f) = 0.69 ± 0.21 (rms) and -0.08 ± 0.23 (FWHM)
Derived by comparing dynamical M_BH to virial products across the sample of 38 objects.

axioms (1)

domain assumption BLR geometry is best described as a thick disk with rotational, inflowing, and outflowing components viewed at intermediate inclinations
This is stated as the best-fit description from the dynamical modeling of the eight objects.

pith-pipeline@v0.9.0 · 5737 in / 1511 out tokens · 131215 ms · 2026-05-08T02:29:23.121093+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

3 extracted references · 3 canonical work pages

[1]

2024, Nature, 627, 281 Bao, D.-W., Brotherton, M

Abuter, R., Allouche, F., Amorim, A., et al. 2024, Nature, 627, 281 Bao, D.-W., Brotherton, M. S., Du, P., et al. 2022, ApJS, 262, 14 Barth, A. J., Bennert, V. N., Canalizo, G., et al. 2015, ApJS, 217, 26 Batiste, M., Bentz, M. C., Raimundo, S. I., Vestergaard, M., & Onken, C. A. 2017, ApJL, 838, L10 Bellm, E. C., Kulkarni, S. R., Graham, M. J., et al. 20...

work page arXiv 2024
[2]

Seoul National University AGN Monitoring Project. V. Velocity-resolved H<i>β</i> Reverberation Mapping and Evidence of Kinematics Evolution

https://dx.doi.org/10.3847/1538-4357/adbca5 Williams, P. R., & Treu, T. 2022, ApJ, 935, 128 Williams, P. R., Pancoast, A., Treu, T., et al. 2018, ApJ, 866, 75 —. 2020, ApJ, 902, 74 21 Winkel, N., Bennert, V. N., Remigio, R. P., et al. 2025, ApJ, 978, 115 Woo, J.-H., Schulze, A., Park, D., et al. 2013, ApJ, 772, 49 Woo, J.-H., Treu, T., Malkan, M. A., & Bl...

work page doi:10.3847/1538-4357/adbca5 2022
[3]

2022, MNRAS, 517, 2659 Wu, Q., & Shen, Y

http://dx.doi.org/10.3847/1538-4357/ad132f Wu, J., Shen, Y., Jiang, L., et al. 2022, MNRAS, 517, 2659 Wu, Q., & Shen, Y. 2022, ApJS, 263, 42 Xiao, M., Du, P., Horne, K., et al. 2018a, ApJ, 864, 109 Xiao, M., Du, P., Lu, K.-K., et al. 2018b, ApJL, 865, L8 Yang, S., Du, P., & Wang, J.-M. 2024, ApJS, 274, 24 Yu, L.-M., Bian, W.-H., Wang, C., Zhao, B.-X., & G...

work page doi:10.3847/1538-4357/ad132f 2022

[1] [1]

2024, Nature, 627, 281 Bao, D.-W., Brotherton, M

Abuter, R., Allouche, F., Amorim, A., et al. 2024, Nature, 627, 281 Bao, D.-W., Brotherton, M. S., Du, P., et al. 2022, ApJS, 262, 14 Barth, A. J., Bennert, V. N., Canalizo, G., et al. 2015, ApJS, 217, 26 Batiste, M., Bentz, M. C., Raimundo, S. I., Vestergaard, M., & Onken, C. A. 2017, ApJL, 838, L10 Bellm, E. C., Kulkarni, S. R., Graham, M. J., et al. 20...

work page arXiv 2024

[2] [2]

Seoul National University AGN Monitoring Project. V. Velocity-resolved H<i>β</i> Reverberation Mapping and Evidence of Kinematics Evolution

https://dx.doi.org/10.3847/1538-4357/adbca5 Williams, P. R., & Treu, T. 2022, ApJ, 935, 128 Williams, P. R., Pancoast, A., Treu, T., et al. 2018, ApJ, 866, 75 —. 2020, ApJ, 902, 74 21 Winkel, N., Bennert, V. N., Remigio, R. P., et al. 2025, ApJ, 978, 115 Woo, J.-H., Schulze, A., Park, D., et al. 2013, ApJ, 772, 49 Woo, J.-H., Treu, T., Malkan, M. A., & Bl...

work page doi:10.3847/1538-4357/adbca5 2022

[3] [3]

2022, MNRAS, 517, 2659 Wu, Q., & Shen, Y

http://dx.doi.org/10.3847/1538-4357/ad132f Wu, J., Shen, Y., Jiang, L., et al. 2022, MNRAS, 517, 2659 Wu, Q., & Shen, Y. 2022, ApJS, 263, 42 Xiao, M., Du, P., Horne, K., et al. 2018a, ApJ, 864, 109 Xiao, M., Du, P., Lu, K.-K., et al. 2018b, ApJL, 865, L8 Yang, S., Du, P., & Wang, J.-M. 2024, ApJS, 274, 24 Yu, L.-M., Bian, W.-H., Wang, C., Zhao, B.-X., & G...

work page doi:10.3847/1538-4357/ad132f 2022