How did they actually take this picture? (Very Long Baseline Interferometry)

The Event Horizon Telescope’s black-hole images are possible only because Earth-based radio observatories act together like an Earth-sized telescope, using very long baseline interferometry to sharpen radio-wave measurements to the scale of a black hole’s “shadow.” Sagittarius A*—the supermassive black hole at the Milky Way’s center—was especially hard: it sits behind thick dust and gas that blocks visible light, it is over 1,000 times smaller than M87* (so it appears only slightly larger from Earth), and its accretion flow changes on minute timescales. Yet the collaboration still produced a ring-like silhouette by observing at radio wavelengths of 1.3 millimeters, where the surrounding plasma is detectable even though the black hole itself emits no light.

The core technical trick starts with angular resolution, the ability to distinguish tiny structure on the sky. A single radio dish can’t resolve the ring structure of a black hole because its resolution is too coarse; as a telescope scans across the target, it would blur together signals from different parts of the black hole. Shorter wavelengths could, in principle, improve resolution, but Earth’s atmosphere and the black hole’s environment block those options. Instead, resolution is boosted by effectively increasing the telescope’s diameter. The Event Horizon Telescope accomplishes this not with one giant dish, but by linking widely separated dishes across the planet—up to distances comparable to Earth’s diameter—so their combined signals reproduce the interference pattern an Earth-sized instrument would generate.

Each observatory records the incoming radio signal locally along with extremely precise timing (to the femtosecond). The data can’t be combined in real time because the signals arrive at different places, so petabytes of recordings are shipped to central processing sites. The imaging step then relies on interference fringes: pairs of telescopes separated by different distances and orientations produce different bright-and-dark patterns depending on the wavefront’s phase and the source’s position. By collecting many such patterns across a global network, the system reconstructs a high-resolution image that no single telescope could produce.

What the “black shadow” corresponds to is subtler than the event horizon itself. Space-time curvature bends light rays, so photons that skim near the black hole can either fall in or escape depending on their approach angle. The resulting dark region is set by the photon sphere: light can orbit at about 1.5 Schwarzschild radii, but that orbit is unstable. Rays that just graze the photon sphere form a shadow whose diameter is about 2.6 times the event horizon’s radius. In that sense, the center of the shadow maps to the event horizon, while the surrounding darkness reflects how many light paths are swallowed.

The bright ring and asymmetries come from the accretion disk’s hot, fast-moving plasma. Matter orbits at significant fractions of the speed of light, so relativistic (Doppler) beaming makes the side moving toward Earth brighter than the side moving away. Depending on viewing angle, gravitational lensing can also reveal the “backside” of the accretion disk and even multiple, nested images of the event horizon—light that loops around the black hole before reaching the telescopes. The final picture is therefore not a direct photograph of a black hole, but a map of how curved space-time and relativistic plasma dynamics shape the paths and brightness of incoming radio-emitting light.