First M87 Event Horizon Telescope Results. V. Physical Origin of the Asymmetric Ring

When the Event Horizon Telescope Collaboration released its 2017 image of M87's galactic core, the picture that arrived was unlike any radio map before it: a bright, asymmetric ring of emission encircling a deep central darkness. The ring had an angular diameter of roughly 40 microarcseconds, which is 40 millionths of an arcsecond — the apparent size of an orange on the Moon. The total flux at 1.3 millimeters was about 0.5 Jansky, and the peak brightness temperature hit around 6 billion Kelvin. Anchoring all of this was an independently measured central mass of about 6.5 billion solar masses, sitting 16.9 megaparsecs away. The image was striking. But the collaboration's fifth paper pushed past the visual. The real question wasn't "did we see a black hole?" It was: what is the physical origin of this specific asymmetric ring? The ring has two causes working in concert. The first is gravitational lensing. Lines of sight that graze the black hole linger near unstable photon orbits — paths where light can actually circle the hole before escaping. Those sightlines sample long stretches of hot, synchrotron-emitting plasma, making them comparatively bright. The result is the photon ring encircling a central darkness, what the collaboration calls the black hole shadow. The second cause is Doppler beaming. The plasma orbiting the black hole is moving relativistically. The side swinging toward us gets boosted in brightness, while the side swinging away gets dimmed. That's why the southern arc in the M87 image is brighter than the northern arc — south is the approaching side, given the orientation of M87's large-scale jet at a viewing angle of about 17 degrees. The ring radius tracks black hole mass. The brightness asymmetry tracks spin and viewing geometry. Both are expected to be stable features in future observations. To test whether this physical picture actually explains the data, the collaboration didn't just try a handful of models. They built a library. The strategy was to simulate the plasma physics near the black hole using general relativistic magnetohydrodynamics, which is abbreviated as GRMHD for short. This approach numerically evolves plasma in the warped spacetime of a spinning Kerr black hole. A weakly magnetized torus is set orbiting the hole, driven turbulent by the magnetorotational instability, and then evolved until the accretion flow settles into a statistically relaxed state. The library contains 43 distinct GRMHD simulations, occupying 23 terabytes of disk space, run across multiple independent codes to verify consistency. Two qualitatively different accretion regimes anchor the library. SANE models — which stands for Standard and Normal Evolution — accumulate modest magnetic flux near the horizon. MAD models — which stands for Magnetically Arrested Disks — accumulate so much magnetic flux that it can temporarily halt the infalling plasma. The physics at the horizon is qualitatively different between them, and the jet power they produce differs dramatically. Spin was sampled widely: SANE runs at nine values from nearly retrograde to nearly prograde, and MAD at six. To handle electron physics — a genuine uncertainty — the team used a one-parameter prescription called R-high, which sets how hot the electrons are in the dense disk midplane relative to the ions. Six values of R-high were tested, from 1 to 160. Each GRMHD snapshot was then post-processed with general relativistic ray-tracing codes — ipole, RAPTOR, and BHOSS — to generate synthetic radio images. The resulting image library contains around 60,000 synthetic images. This is the tool the collaboration used to interrogate the data. The filtering happened in three stages, each one a separate observational fence. The first was the EHT image itself: does the time-averaged model image reproduce the observed 1.3 millimeter ring? That cut used a statistical image-scoring tool called THEMIS-AIS to compare synthetic visibilities against the data. The second was X-ray emission. Chandra and NuSTAR observed M87 simultaneously with the Event Horizon Telescope in April 2017, measuring a 2 to 10 kiloelectronvolt luminosity of 4.4 times ten to the fortieth ergs per second. Any model producing average X-ray emission above that observational value was rejected. The most X-ray-luminous models were SANE configurations with small R-high. Low R-high means hotter electrons in the disk, which means more inverse-Compton scattering and more X-rays. One SANE model with R-high equal to 10 produced an average X-ray luminosity of 4.2 times ten to the forty-second ergs per second — nearly a hundred times the observed cap. Those models were out. The third filter was jet power, and this one did the most damage to the model library. M87's jet is one of the most powerful in the known universe. Conservative estimates put the minimum jet power at ten to the forty-second ergs per second. Every non-spinning model in the library — every simulation with a spin parameter called a-star equal to zero — failed this test. Not marginally. The most powerful non-spinning MAD case produced a total outflow power of only 3.7 times ten to the forty-first ergs per second, and its jet power was consistent with zero. Zero-spin black holes simply cannot drive M87's jet. The models that did pass all three filters were predominantly high-spin, magnetically arrested configurations. And the mechanism supplying their jet power matched the Blandford-Znajek process: magnetic fields threading the spinning black hole's horizon extract rotational energy and channel it into a relativistic Poynting flux along the polar axis. The simulations showed that the jet power scales with spin and magnetic flux in the way Blandford-Znajek predicts, though the coefficient varies — MAD models produce larger funnels and more efficient extraction than SANE models, with coefficients ranging from about 0.15 to 0.83 of the analytic estimate. If you tightened the jet power requirement to three times ten to the forty-second ergs per second, only models with a spin magnitude above 0.94 and R-high of at least 10 would survive. The conclusion was unambiguous: the black hole at the center of M87 is spinning, its spin energy is powering the jet, and the spin vector — if aligned with the jet — points away from Earth. That said, the collaboration was careful about what remains genuinely uncertain. The largest open question is electron physics. The R-high prescription is a coarse proxy for a complicated set of kinetic processes, including particle heating, radiative cooling, Coulomb coupling, and possible pair production. Radiation GRMHD calculations — which are far more expensive to run — show that for SANE models, the electron temperature distribution they produce resembles the R-high equals 10 case, while for MAD models the radiation-GRMHD temperature is roughly three times lower than the R-high equals 20 case. The one-parameter model compresses all of that into a single dial. The team also assumed a purely thermal electron energy distribution throughout. Nonthermal electron distributions — those with a power-law tail of high-energy particles — can reproduce the observed spectral energy distribution too, and they tend to increase emission per unit accretion rate, implying lower accretion rates than the values quoted above. The electron physics is a real uncertainty, not just academic hedging. There is also the question of whether the compact object is actually a black hole at all. The collaboration considered three categories of alternatives: modified black hole solutions within general relativity, black holes in alternative gravity theories, and horizonless mimickers — objects like boson stars, wormholes, gravastars, and superspinars. Many of these produce shadows qualitatively similar to Kerr and cannot be ruled out by the 2017 image alone. But some extremes are already excluded. Superspinars and certain naked singularity models with a filled central disk are inconsistent with the Event Horizon Telescope 2017 data. Horizonless objects that match the static ring image still predict different accretion dynamics and variability — handles that future observations can grab. The roadmap from here involves polarization and higher-frequency imaging. The Event Horizon Telescope took polarized data in 2017, and the collaboration expects polarization comparisons to sharply constrain the model library in ways intensity maps cannot. Future campaigns at 345 gigahertz — compared to the 230 gigahertz of the 2017 observations — will reduce optical depth, make the ring narrower and better defined, and reduce sensitivity to uncertain plasma prescriptions. Multi-epoch observations, separated by more than the model decorrelation time of roughly two weeks, will further multiply the discriminating power. These observations push into the most extreme gravitational regime ever imaged — a direct test of general relativity where the spacetime curvature is as strong as it gets. The 2017 result established that the image is consistent with a spinning Kerr black hole. The work ahead is to find out just how tight that consistency really is. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.