How Scientists Detect Black Holes They Cannot Directly See

Seeing the Invisible

On April 10, 2019, the Event Horizon Telescope Collaboration released the first image of a black hole — a blurry orange ring surrounding a dark central region 6.5 billion times the mass of the Sun, located 55 million light-years away in the galaxy M87. The image required eight radio telescope arrays across four continents, a petabyte of hard drives, and algorithms developed by a 29-year-old MIT graduate student named Katie Bouman. What it showed was not the black hole itself — nothing can escape from inside the event horizon — but the shadow it cast on the glowing gas swirling around it.

This image was the culmination of a century's work developing indirect detection methods, because direct observation of a black hole is by definition impossible. No electromagnetic radiation of any kind — no visible light, no X-rays, no radio waves — escapes from within the event horizon. Everything astronomers know about black holes comes from observing how they affect their surroundings: distorting spacetime, accelerating matter to near light speed, flinging radiation outward from gas that falls toward them, and sending ripples through the fabric of space when two of them collide.

Method 1: X-Ray Binaries

The first strong evidence for stellar-mass black holes came from X-ray astronomy in the 1960s and 70s. When a black hole orbits a normal star, it can pull gas from the companion's outer atmosphere. This gas forms an accretion disk — a rotating disk of superheated plasma spiraling inward. As gas falls into the disk's inner regions, it reaches temperatures of 10 million Kelvin or more, emitting intense X-rays detectable from Earth.

Cygnus X-1, discovered in 1964 by a rocket-borne X-ray detector, was the first strong black hole candidate. Its X-ray luminosity exceeded anything a neutron star could produce, and spectroscopic measurements of its companion star showed orbital velocities indicating the compact object's mass was approximately 21 solar masses — far above the maximum mass for a neutron star (~3 solar masses). Stephen Hawking bet Kip Thorne in 1974 that Cygnus X-1 was not a black hole; Hawking conceded in 1990.

Method 2: Stellar Orbits at the Galactic Center

The most direct mass measurement of a black hole comes from tracking stars that orbit very close to it. At the center of the Milky Way lies a compact radio source called Sagittarius A* (Sgr A*), suspected to be a supermassive black hole. Beginning in the 1990s, astronomer Reinhard Genzel at the Max Planck Institute and Andrea Ghez at UCLA independently began tracking individual stars near Sgr A* using adaptive optics on large ground-based telescopes.

Over decades of observation, they mapped the complete orbits of stars within the innermost 0.04 light-years of the galactic center — including a star called S2 with an orbital period of just 16 years and a closest approach to Sgr A* of about 120 AU (roughly 3 times Pluto's distance from the Sun). Applying Kepler's laws to these orbits yielded an enclosed mass of approximately 4.1 million solar masses concentrated in a region smaller than our solar system. Only a black hole can contain that much mass in that small a volume. Genzel and Ghez shared the 2020 Nobel Prize in Physics for this work.

Method 3: Gravitational Waves

On September 14, 2015, the Laser Interferometer Gravitational-Wave Observatory (LIGO) detected a signal lasting 0.2 seconds — the first direct detection of gravitational waves, and the first observation of two black holes merging. The event, designated GW150914, involved black holes of approximately 36 and 29 solar masses colliding 1.3 billion light-years away. In the final milliseconds, the power output exceeded the total electromagnetic luminosity of all stars in the observable universe combined.

Gravitational wave detection works by interferometry. LIGO's two arms, each 4 km long, measure the fractional change in length caused by a passing gravitational wave. GW150914 stretched and compressed each arm by roughly 10⁻¹⁸ meters — one-thousandth the diameter of a proton — at its peak. By 2026, the LIGO-Virgo-KAGRA network had detected over 90 confirmed gravitational wave events, mostly black hole mergers.

Method 4: Direct Imaging of the Event Horizon Shadow

The Event Horizon Telescope (EHT) is not a single telescope but a global very-long-baseline interferometry (VLBI) array that synthesizes a virtual dish as wide as the Earth. By synchronizing observations at wavelengths of 1.3 mm across eight sites — including Mauna Kea, the South Pole, the Atacama Desert, and European sites — the EHT achieves angular resolution of about 20 microarcseconds. That's enough to resolve a grapefruit on the Moon.

The 2019 M87* image was followed in 2022 by an image of Sgr A*, the Milky Way's own central black hole. Sgr A* was harder to image than M87* despite being much closer because the gas near it moves so fast — orbiting periods of minutes rather than weeks — that the image varies during an observation session, requiring novel imaging algorithms to produce a stable picture from a constantly changing target.

Distinguishing Black Holes From Neutron Stars

Both black holes and neutron stars can appear as X-ray binaries or gravitational wave sources. Several signatures distinguish them:

• Mass gap: neutron stars cannot exceed approximately 2.5–3 solar masses due to degeneracy pressure limits. Objects above this threshold in compact binary systems are almost certainly black holes.
• Type I X-ray bursts: neutron stars have hard surfaces; accreted hydrogen ignites in periodic thermonuclear flashes detectable as X-ray bursts. Black holes have no surface; infalling material passes the event horizon without producing bursts.
• Gravitational wave ringdown: after two black holes merge, the resulting merged black hole oscillates and radiates gravitational waves at frequencies predicted precisely by general relativity. This quasinormal mode pattern is unique to black holes and has been detected in several LIGO events.

What Has Not Been Detected Yet

Despite hundreds of confirmed stellar-mass and supermassive black holes, intermediate-mass black holes (IMBHs) — in the range 100 to 100,000 solar masses — remain elusive. They are expected to exist in globular clusters and dwarf galaxies, but definitive confirmations are few. Some gravitational wave events (GW190521, involving a final mass of ~142 solar masses) may have produced IMBHs, but the merging objects were themselves anomalously massive.

Primordial black holes — hypothetical objects formed in the early universe — have been proposed as dark matter candidates. Gravitational microlensing surveys have placed constraints on their abundance but haven't confirmed or ruled them out in the relevant mass ranges. The Pulsar Timing Array results from 2023, showing a gravitational wave background consistent with supermassive black hole binary inspirals throughout cosmic history, opened yet another observational window into the most extreme objects in the universe.