The Largest Radio Telescope Arrays

China's Five-hundred-meter Aperture Spherical Telescope (FAST), completed in 2016 in a natural karst depression in Guizhou province, holds the record for the largest single-dish radio telescope on Earth, with a dish 500 meters (1,640 feet) in diameter. That is nearly twice the diameter of the previous record-holder, the Arecibo Observatory in Puerto Rico, which collapsed in December 2020 after cable failures sent its 900-ton instrument platform crashing into the dish below. But single dishes are not where the real story of "biggest" radio telescopes ends. The largest effective radio telescopes in astronomy are not single structures at all — they are arrays, networks of smaller antennas spread across deserts, continents, and eventually low Earth orbit, working together to simulate the resolving power of one impossibly enormous dish. Understanding why requires understanding a trick of physics called interferometry, and once you see how it works, the difference between "biggest dish" and "biggest telescope" stops being a technicality and starts being the whole point.

Why bigger dishes and more antennas both matter

A telescope's ability to resolve fine detail, called angular resolution, improves with the diameter of its collecting area relative to the wavelength of light it observes. Radio waves are enormously long compared to visible light. A typical radio observation might use wavelengths from a few millimeters to several meters, while visible light wavelengths are measured in hundreds of nanometers, tens of thousands of times shorter. That mismatch is why radio telescopes must be huge, hundreds of meters or, more often, spread across kilometers or continents, just to achieve the kind of resolving power an optical telescope gets from a mirror a few meters wide.

FAST's 500-meter dish gives it enormous raw sensitivity, meaning its ability to detect very faint incoming radio signals, because a larger collecting area intercepts more of the radio energy arriving from a distant source. That sensitivity is why FAST has become a leading instrument for pulsar detection: since beginning full operations, the observatory has been credited by China's National Astronomical Observatories with discovering more than 800 new pulsars, rapidly spinning neutron stars that emit regular beams of radio waves, as of announcements through 2023 (National Astronomical Observatories of China, 2023). But a single dish, no matter how large, has a hard resolution ceiling set by its own physical diameter. To see genuinely fine detail — to resolve the swirl of gas around a black hole's event horizon, for instance — astronomers need something no single dish, not even one the size of a small country, could ever achieve alone. They need arrays.

FAST's dish is built from 4,450 individual triangular panels whose shape can be actively adjusted to focus incoming radio waves on a suspended receiver cabin, effectively turning a fixed bowl in the ground into a steerable instrument (National Astronomical Observatories of China, 2016).

How interferometry turns many small dishes into one giant telescope

Radio interferometry combines the signals from multiple separate antennas, then uses the tiny time delays between when each antenna receives the same incoming wavefront to mathematically reconstruct an image with the resolution of a single telescope as large as the maximum distance between the antennas, a distance astronomers call the baseline. Think of it like listening to a car alarm with two ears instead of one: your brain compares the split-second delay between when the sound hits your left ear versus your right, and uses that tiny timing difference to figure out precisely where the sound is coming from. A radio interferometer does the same thing electronically, except instead of two ears a few centimeters apart, it might use dozens of antennas spread across kilometers, or observatories spread across an entire hemisphere.

This means an array spread across kilometers, or even across continents, can achieve resolution far beyond what any single dish could reach on its own, no matter how enormous that dish might be. The tradeoff is that spreading antennas out sacrifices some sensitivity per unit of collecting area unless you also add more antennas, which is why the biggest interferometric projects combine both strategies: many antennas, arranged over the widest baselines the science requires.

The Atacama Large Millimeter/submillimeter Array (ALMA), sited at 5,000 meters (16,400 feet) elevation on the Chajnantor plateau in Chile's Atacama Desert, uses 66 individual antennas, 54 with 12-meter dishes and 12 with 7-meter dishes, that can be physically repositioned by specialized transporter vehicles across baselines ranging from about 150 meters up to 16 kilometers (10 miles) (ALMA Observatory / European Southern Observatory). Moving the antennas closer together improves sensitivity to large, faint structures like molecular clouds; spreading them to the maximum baseline sharpens resolution for imaging fine detail, such as the disks of gas and dust around young stars where planets are forming. ALMA operates at millimeter and submillimeter wavelengths, shorter than most radio arrays, which lets it study the cold, dusty regions of star and planet formation that are largely invisible in ordinary light.

How does an interferometer's baseline determine its resolution?

The maximum resolution of an interferometric array corresponds to a telescope with a single dish as wide as the array's longest baseline, the greatest distance between any two of its antennas. A wider baseline sharpens the image; more antennas at varied baselines fill in detail and reduce the empty gaps between signals, similar to adding more pixels to a photograph the wider a lens is separating two viewpoints.

The Very Large Array: America's reconfigurable radio observatory

The Karl G. Jansky Very Large Array (VLA), on the Plains of San Agustin in New Mexico, has been one of the world's most productive radio observatories since it began full operations in 1980. It consists of 27 dish antennas, each 25 meters (82 feet) in diameter, arranged in a Y-shaped track configuration that can be physically reconfigured into four different layouts, ranging from a compact "D" configuration spanning about 1 kilometer to an extended "A" configuration with baselines up to 36 kilometers (22 miles) (National Radio Astronomy Observatory). Astronomers choose the configuration depending on whether a project needs wide-field sensitivity to faint, extended structures or sharp resolution on a compact source.

The VLA's fame extends well beyond astrophysics circles. It appeared as a filming location in the 1997 movie "Contact," based on Carl Sagan's novel, and its dramatic Y-shaped array of white dishes against the New Mexico high desert has become one of the most recognizable images of an astronomical observatory in popular culture. Scientifically, the VLA has been central to work ranging from mapping the structure of the Milky Way to tracking radio emission from supernova remnants and detecting the first exoplanet imaged around a star other than the sun's own solar neighbors.

The Square Kilometre Array: the largest radio observatory ever attempted

The Square Kilometre Array Observatory (SKAO) is an intergovernmental organization, headquartered in the United Kingdom, building what will become the largest radio telescope network ever constructed, split between two sites on opposite sides of the Indian Ocean. SKA-Mid, under construction in South Africa's Northern Cape province, will eventually combine roughly 197 dish antennas in its full design baseline, including the 64 dishes of the existing MeerKAT array that South Africa built as a precursor instrument, spread across baselines reaching roughly 150 kilometers (93 miles) (SKA Observatory). SKA-Low, under construction in Western Australia on Wajarri Yamaji Country, will use a different design entirely: rather than dishes, it will rely on hundreds of low-frequency antenna stations, each a cluster of simple dipole antennas with no moving parts, that electronically combine their signals to reach the observatory's full design target of 512 stations (SKA Observatory).

The SKA's scale is deliberately international. The project is funded and governed by a consortium of member countries that as of the mid-2020s includes Australia, China, France, India, Italy, the Netherlands, Portugal, South Africa, Spain, Switzerland, and the United Kingdom, with additional countries participating at earlier partnership stages (SKA Observatory, Council membership listings). In January 2026, SKAO announced that two of the SKA-Mid dishes in South Africa achieved "first fringes," meaning they successfully combined a real astronomical signal as an interferometric pair for the first time, a milestone that confirmed the telescope's core electronics and correlator systems were functioning together as designed (SKA Observatory, 2026). Full science operations are being phased in through the late 2020s and into the 2030s as more dishes and stations come online.

The Square Kilometre Array does not get its name from a single square kilometer of dishes clustered together. The name refers to the combined collecting area the observatory aims for across its full design baseline, spread over antennas separated by up to roughly 150 kilometers, not a single physical footprint.

What scientific goals justify building the SKA at such enormous scale?

Two flagship goals drive the SKA's design. First, mapping neutral hydrogen gas across cosmic time, including the faint 21-centimeter radio signal emitted by hydrogen atoms during the "cosmic dawn," the period when the universe's first stars and galaxies reionized the surrounding gas roughly 400 million to 1 billion years after the Big Bang. Second, running a large-scale pulsar timing array, using the ultra-precise, clock-like regularity of dozens of millisecond pulsars scattered across the galaxy to detect nanohertz-frequency gravitational waves, ripples in spacetime with wavelengths of light-years, generated primarily by pairs of merging supermassive black holes.

The Event Horizon Telescope: interferometry at planetary scale

The most extreme application of the interferometry principle already exists and has already delivered results: the Event Horizon Telescope (EHT), a collaboration that links radio observatories on different continents using very-long-baseline interferometry (VLBI), effectively creating a virtual telescope with a baseline as large as Earth itself. In April 2019, the Event Horizon Telescope Collaboration published the first-ever direct image of a black hole's event horizon, the boundary beyond which nothing, not even light, can escape, showing the supermassive black hole at the center of the galaxy M87, roughly 55 million light-years from Earth and with a mass estimated at 6.5 billion times that of the sun (Event Horizon Telescope Collaboration, Astrophysical Journal Letters, 2019). In May 2022, the same collaboration released a second image, this time of Sagittarius A*, the far smaller and closer supermassive black hole at the center of our own Milky Way, with a mass estimated at roughly 4 million solar masses (Event Horizon Telescope Collaboration, Astrophysical Journal Letters, 2022).

Producing those images required linking observatories including ALMA in Chile, the South Pole Telescope, the IRAM 30-meter telescope in Spain, the James Clerk Maxwell Telescope and Submillimeter Array in Hawaii, and several others, all observing the same target simultaneously while synchronized by extraordinarily precise atomic clocks, hydrogen maser oscillators accurate to roughly one second in tens of millions of years. Rather than physically wiring the observatories together, which is impossible across such distances, each site records its own data along with the exact atomic-clock timestamp of every observation. The recorded data, often amounting to petabytes per observing campaign, is then physically shipped or transmitted to a central supercomputing facility, where specialized software called a correlator combines the datasets after the fact, using the timestamps to align signals as if they had all arrived at one giant, Earth-sized dish.

Comparing the major arrays and dishes

Facility	Scale	Type	Location
FAST	500 m dish diameter	Single dish, largest in the world	Guizhou, China
Arecibo Observatory (collapsed 2020)	305 m dish diameter	Single dish, former record holder	Puerto Rico, USA
ALMA	66 antennas, baselines up to 16 km	Interferometric array	Atacama Desert, Chile
Very Large Array	27 antennas, baselines up to 36 km	Interferometric array	New Mexico, USA
Square Kilometre Array (SKA-Mid + SKA-Low, full design)	Up to 197 dishes + 512 antenna stations	Interferometric array, two continents	South Africa and Australia
Event Horizon Telescope	Global network of observatories	Very-long-baseline interferometry, Earth-scale baseline	Multiple continents

Raw dish size and effective resolving power are different records — FAST wins the first, arrays win the second by orders of magnitude.

What people get wrong about "biggest" radio telescopes

A common misconception is that a bigger dish always means better science, full stop. In practice, FAST's enormous single dish provides excellent raw sensitivity for detecting faint or distant radio sources, like pulsars deep in the galactic plane, but arrays like ALMA, the VLA, and the EHT achieve far sharper resolution for imaging fine structural detail. That is because resolution scales with the maximum separation between antennas, not with the physical size of any single dish. A dish the size of a football stadium and an array of small dishes spread across a desert can, in principle, deliver very different pictures of the same object: one tells you how much energy is arriving, the other tells you exactly where, in fine detail, it is coming from.

Another mistake is assuming an array's antennas need to be physically connected by cable to function together. Very-long-baseline interferometry, the technique behind the Event Horizon Telescope, instead uses extremely precise atomic clocks at each observatory to timestamp incoming data independently. That data is then combined and processed afterward at a central facility, allowing antennas thousands of kilometers apart, with no physical connection whatsoever, to function as a single instrument. There is effectively no upper limit on baseline length imposed by cabling, which is why VLBI networks like the EHT can span entire continents, and why proposed future projects have discussed placing radio antennas in orbit to create baselines even larger than Earth's own diameter.

A third, more subtle misconception is treating "largest telescope" as a single, uncontested title. FAST is unambiguously the largest single dish. The VLA and ALMA are large by antenna count and physical footprint but modest compared to the SKA's eventual scale. The SKA, once its full design baseline is realized, will be the largest radio observatory ever built by combined collecting area and antenna count, spanning two continents under one scientific governance structure. And the EHT, despite using comparatively few individual dishes, achieves the single sharpest angular resolution of any telescope on the list, because its effective baseline is the diameter of the planet. Each answer to "which is biggest" depends entirely on which property you are measuring: dish diameter, total collecting area, antenna count, or baseline length.

Why can't astronomers just build one dish as big as an entire array's baseline?

Physical dishes require rigid support structures whose costs and engineering demands grow far faster than their diameter. A dish scaled to the size of the Event Horizon Telescope's effective Earth-diameter baseline is not merely impractical, it is physically impossible to build and support under gravity. Interferometry sidesteps that limit entirely by combining many modest, buildable dishes instead of attempting one impossibly large structure.

The takeaway

The largest radio telescope in raw dish size is China's FAST, at 500 meters across, but the largest effective telescopes in terms of image resolution are interferometric arrays: ALMA and the Very Large Array today, the Square Kilometre Array as it comes fully online through the 2030s, and the Event Horizon Telescope, which already combines signals from observatories spread across the globe to simulate the resolving power of a single, planet-sized dish. None of these facilities compete for the exact same title. FAST wins on raw light-gathering power for faint, distant sources. The EHT wins on sheer image sharpness. The SKA, once complete, will win on total collecting area and scientific breadth, from mapping hydrogen across cosmic history to listening for gravitational waves in the timing of pulsars scattered across the galaxy. The real record isn't a single number. It's the trick, shared by every array on this list, of making many modest instruments think and see as one.

---

Read next:

• The Largest Telescopes on Earth
• The Most Powerful Particle Accelerators
• The Fastest Human-Made Objects Ever

FAQ

Q: What is the largest single-dish radio telescope in the world?

China's Five-hundred-meter Aperture Spherical Telescope (FAST), completed in 2016 in Guizhou province, with a dish 500 meters (1,640 feet) in diameter. It has been credited with discovering more than 800 new pulsars since beginning full operations (National Astronomical Observatories of China, 2023).

Q: What happened to the Arecibo Observatory?

Arecibo, previously the record-holding single-dish radio telescope at 305 meters in diameter, suffered cable failures that caused its 900-ton instrument platform to collapse into the dish in December 2020, ending its operational history after more than 50 years of service.

Q: How does the Event Horizon Telescope achieve such extreme resolution?

It links radio observatories on different continents using very-long-baseline interferometry, combining their data with extremely precise atomic clock timestamps to simulate a telescope with a baseline as large as Earth itself. That technique produced the first direct images of black hole event horizons, of M87's black hole in 2019 and of the Milky Way's own Sagittarius A* in 2022.

Q: Is a bigger single radio dish always better than an array of smaller antennas?

Not for every purpose. A large single dish like FAST offers strong raw sensitivity for detecting faint signals across wide areas of sky, while arrays like ALMA and the Very Large Array achieve much finer image resolution, since resolution depends on the maximum distance between antennas rather than the size of any one dish.

Q: What makes the Square Kilometre Array different from existing radio arrays like ALMA or the VLA?

The SKA is designed to be far larger in both antenna count and total collecting area, split between two design types across two continents: up to roughly 197 dishes in South Africa's SKA-Mid and up to 512 low-frequency antenna stations in Australia's SKA-Low. Its flagship goals include mapping neutral hydrogen from the universe's earliest galaxies and running a pulsar timing array to detect gravitational waves from merging supermassive black holes (SKA Observatory).