Physical origin of the minute-scale horizon-scale dynamics detected in Event Horizon Telescope observations of Sagittarius A*

Bayesian dynamical imaging of the Event Horizon Telescope's (EHT) April 6, 2017 observations of Sagittarius A* (Sgr A*) recovered a brightness feature near the black hole that propagated clockwise on the sky between roughly 10:30 and 13:30 UTC at one-minute time resolution (Knollmüller, Arras and Enßlin 2023). The origin of this feature is not established: geometric ray-traced modeling yields an emission radius ($\approx 4.7\,GM/c^2$, inside the Schwarzschild innermost stable circular orbit) that is mutually inconsistent with the observed pattern speed ($\approx 2^\circ$/min, implying three-hour dynamics far slower than orbital motion at that radius), so a localized orbiting emitter, apparent motion from large-scale structural evolution of the accretion flow or jet, and artifacts of modeling simplifications, unmodeled refractive interstellar scattering, or residual instrumental effects all remain viable. The problem is to determine which explanation the data support, taking into account the effects the source analysis left untreated (refractive scattering, station-level systematics, posterior multimodality, physically motivated emission models), and to report physical parameters only at the precision this discrimination justifies. No published follow-up resolves this as of 2026.

Background

Sagittarius A* (Sgr A*) is the supermassive black hole ($\approx 4\times 10^6$ solar masses) at the center of the Milky Way. The Event Horizon Telescope (EHT) is a global very long baseline interferometric (VLBI) array operating at 1.3 mm wavelength, in which widely separated radio dishes measure visibilities — samples of the Fourier transform of the sky brightness distribution. Because Sgr A*'s emission varies on timescales of minutes, comparable to the orbital time near the black hole, the standard VLBI assumption of a static source during an observation fails; Earth-rotation aperture synthesis, which normally fills Fourier space as the Earth turns, cannot be used. The EHT Collaboration's 2022 first results therefore imaged a time-averaged ring and mitigated variability by averaging over days and methods rather than reconstructing the dynamics (Event Horizon Telescope Collaboration 2022).

Knollmüller, Arras and Enßlin (2023) re-analyzed the April 6 and 7, 2017 EHT data with the Bayesian dynamical-imaging framework Resolve, which models the log-brightness as an adaptive Gaussian process (a stochastic prior favoring smooth spatio-temporal evolution) and approximates the posterior over tens of millions of image parameters with metric Gaussian variational inference (MGVI), an optimization-based uni-modal approximation to the posterior distribution. On April 6 they recovered a bright feature that emerged in the southwest near 10:30 UTC and propagated clockwise until dissipating near 13:30 UTC; April 7 showed no coherent motion. They then fit this isolated feature with a deliberately simple geometric model: a hot spot (a transient localized bright emission region) on a flat disk, ray-traced through curved spacetime, with linear angular evolution, analyzed by dynamic nested sampling (a Bayesian posterior-exploration algorithm). This produced two modes — an edge-on inclination $\theta_o = 79.6^{\circ}$ (76.5% posterior weight) and a face-on mode $\theta_o = 159.5^{\circ}$ (23.5%), the latter matching the high-inclination, clockwise-sky-motion picture from near-infrared flare astrometry with GRAVITY (a VLTI infrared interferometer; GRAVITY Collaboration 2018) and polarimetric ALMA observations (Wielgus et al. 2022).

The geometric fit, however, is internally inconsistent: the characteristic radius $r_c \approx 4.7M$ (with $M \equiv GM/c^2$ the gravitational radius) lies inside the innermost stable circular orbit (ISCO; $6M$ for a non-rotating black hole), implying prograde spin $|a| \geq 0.37$ if no emission comes from below the ISCO, while the measured angular velocity of about $2^\circ$/min implies a roughly three-hour period with pattern speed $0.056c$ — far below the $0.41c$ expected at the ISCO, and instead matching the $0.23c$ expected at $r_c \approx 19.2M$, which the geometric fit excludes. So either the radius or the angular velocity is unreliable, and the authors lean away from the orbiting-hot-spot interpretation also because no flare appears in the concurrent light curves. Three further limitations bear on any interpretation: only diffractive interstellar scattering (the deterministic blurring kernel of Johnson et al. 2018, caused by turbulent plasma along the line of sight) was forward-modeled, while refractive scattering (the stochastic image substructure the same turbulence produces) was not, though the authors judge a static scattering screen an unlikely source of the observed dynamic feature; residual instrumental and calibration effects — e.g. the changing number of participating antennas — were not conclusively excluded, though the feature appears during optimal, stable coverage; and the uni-modal MGVI posterior cannot fully characterize the known multi-modality of the imaging problem. Interpreting such dynamics physically is normally done against libraries of general relativistic magnetohydrodynamic (GRMHD) simulations — numerical solutions of magnetized plasma accretion combined with relativistic radiative transfer — which the authors did not attempt for this feature. The authors' own forward-looking remarks are narrower than a full research program: the radius/pattern-speed discrepancy "could easily be due to our simplistic model or the extraction procedure and will need further investigation in the future," and if the emission is associated with the jet or lies outside the disk plane, "the dynamics are significantly more complex and we cannot use simple geometric relations to derive conclusions, and a more extensive study based on simulations is required."

Problem Statement

Determine the physical origin of the minute-scale horizon-scale dynamics detected in the Event Horizon Telescope's April 6, 2017 observations of Sagittarius A* — specifically the clockwise-propagating brightness feature recovered between roughly 10:30 and 13:30 UTC by Bayesian dynamical imaging — by discriminating among the three competing explanations left open by the source analysis: (1) a localized orbiting emitter (hot spot) within or near the innermost stable circular orbit; (2) apparent motion arising from large-scale structural evolution of the accretion flow or an associated jet; or (3) an artifact of modeling simplifications, unmodeled refractive interstellar scattering, or residual instrumental/calibration effects. Determine which of the three explanations the April 6, 2017 data support, with the physical parameters (observer inclination, spin, emission radius, pattern speed) reported only to the precision the discrimination justifies. Where the orbiting-emitter explanation (1) is retained, the answer must resolve or explicitly bound the radius-versus-pattern-speed inconsistency of the geometric hot-spot fit ($r_c \approx 4.7\,GM/c^2$ against $\approx 2^\circ$/min); under explanations (2) or (3) it must instead identify which element of that geometric fit fails. The effects the source analysis left untreated — refractive scattering, station-level gain systematics and antenna-coverage changes, posterior multi-modality, and physically motivated emission models in place of its simple geometric hot-spot model — bear on any interpretation of the feature.

References

1. Knollmüller, J., Arras, P., Enßlin, T. (2023). Resolving Horizon-Scale Dynamics of Sagittarius A*. arXiv:2310.16889, https://arxiv.org/abs/2310.16889
2. The Event Horizon Telescope Collaboration, Akiyama, K., Alberdi, A., et al. (2022). First Sagittarius A* Event Horizon Telescope Results. I. The Shadow of the Supermassive Black Hole in the Center of the Milky Way. The Astrophysical Journal Letters 930, L12. DOI 10.3847/2041-8213/ac6674, https://doi.org/10.3847/2041-8213/ac6674
3. The Event Horizon Telescope Collaboration, Akiyama, K., Alberdi, A., et al. (2024). First Sagittarius A* Event Horizon Telescope Results. VIII. Physical Interpretation of the Polarized Ring. The Astrophysical Journal Letters 964, L26. DOI 10.3847/2041-8213/ad2df1, https://doi.org/10.3847/2041-8213/ad2df1
4. Johnson, M. D., Narayan, R., Psaltis, D., et al. (2018). The Scattering and Intrinsic Structure of Sagittarius A* at Radio Wavelengths. The Astrophysical Journal 865, 104. DOI 10.3847/1538-4357/aadcff, https://doi.org/10.3847/1538-4357/aadcff
5. GRAVITY Collaboration, Abuter, R., Amorim, A., et al. (2018). Detection of orbital motions near the last stable circular orbit of the massive black hole SgrA*. Astronomy & Astrophysics 618, L10. DOI 10.1051/0004-6361/201834294, https://doi.org/10.1051/0004-6361/201834294
6. Wielgus, M., Mościbrodzka, M., Vos, J., et al. (2022). Orbital motion near Sagittarius A* — Constraints from polarimetric ALMA observations. Astronomy & Astrophysics 665, L6. DOI 10.1051/0004-6361/202244493 (arXiv:2209.09926), https://doi.org/10.1051/0004-6361/202244493
7. Yfantis, A. I., Wielgus, M., Mościbrodzka, M. A. (2024). Hot spots around Sagittarius A*. Astronomy & Astrophysics 691, A327. DOI 10.1051/0004-6361/202451884, https://doi.org/10.1051/0004-6361/202451884
8. Mus, A., Martí-Vidal, I. (2024). New-generation maximum entropy method: a Lagrangian-based algorithm for dynamic reconstruction of interferometric data. Monthly Notices of the Royal Astronomical Society 528, 5537-5557. DOI 10.1093/mnras/stae234 (arXiv:2401.10177), https://doi.org/10.1093/mnras/stae234
9. Marazzo-Nowicki, B. J., Tiede, P., Chang, D. O., Palumbo, D. C. M., Johnson, M. D. (2026). Measuring the Black Hole and Accretion Parameters of Sagittarius A* from EHT Observations using a Semi-Analytic Model. arXiv:2601.16267, https://arxiv.org/abs/2601.16267
10. Yanagisawa, K., Oka, T., Ariyama, R., et al. (2025). A Face-on Accretion Disk Geometry Revealed by Millimeter-wave Periodicity in Sgr A*. arXiv:2511.02186, https://arxiv.org/abs/2511.02186
11. Kouroshnia, A., Nguyen, K., Ni, C., SaraerToosi, A., Broderick, A. E. (2025). Learning to See: Applying Inverse Recurrent Inference Machines to See through Refractive Scattering. arXiv:2501.14055, https://arxiv.org/abs/2501.14055
12. SaraerToosi, A., Broderick, A. (2025). Validation and Calibration of Semi-Analytical Models for the Event Horizon Telescope Observations of Sagittarius A*. arXiv:2504.18624, https://arxiv.org/abs/2504.18624
13. Albentosa-Ruiz, E., Washington, J. E., Marchili, N., et al. (2026). Full-polarization millimeter wavelength variability of Sagittarius A* during the 2018 EHT campaign. arXiv:2604.10287, https://arxiv.org/abs/2604.10287
14. La Bella, N., Janssen, M., Jeter, B., Müller, H., Van de Berg, B., Pu, H.-Y., Tiede, P., Falcke, H. (2026). Constraining the jet base emission of M87* with past and future Event Horizon Telescope observations. Astronomy & Astrophysics 712, A135. DOI 10.1051/0004-6361/202660438, https://doi.org/10.1051/0004-6361/202660438