The Very Large Array is one of the world’s most productive radio telescopes, a collection of 27 dish antennas arranged in a massive Y-shape across the Plains of San Agustin near Socorro, New Mexico. Since it began full operations in 1980, the facility has contributed to research spanning nearly every corner of astrophysics, from imaging ice hidden in Mercury’s polar craters to detecting the radio glow of colliding neutron stars billions of light-years away. Renamed the Karl G. Jansky Very Large Array after a major electronic overhaul in the 2000s, it remains a workhorse observatory even as planners design its successor.
How Twenty-Seven Dishes Act as One Telescope
A single radio dish, no matter how large, produces a blurry picture of the sky. The sharpness of any telescope depends on the ratio of its aperture to the wavelength of light it collects, and radio waves are far longer than visible light. To get images as sharp as a good optical telescope, a radio dish would need to be tens of kilometers across. Building a single dish that size is impractical, so the VLA uses a technique called interferometry: it combines signals from many smaller antennas spread over a wide area, and the resulting image has the resolution of a dish as large as the distance between the farthest antennas.
Each VLA antenna is 25 meters in diameter and rides on railroad tracks along three arms of a Y, each arm stretching up to about 21 kilometers. Astronomers periodically move the antennas between four standard configurations. The most spread-out arrangement gives the sharpest images, while the most compact arrangement captures large-scale structure that the spread-out configuration misses. A full cycle through all four configurations takes roughly 16 months. The signals from every pair of antennas are combined by a central computer called a correlator, and the combined data are processed into images that can rival optical telescopes in detail.
The original proposal for the VLA, submitted to the National Science Foundation in 1967, called for 36 antennas; the final design was trimmed to 27 to control costs. Even so, the array produces 351 unique antenna pairs, enough to fill in the information needed for high-fidelity images of radio sources across a wide range of angular scales.
The Upgrade That Made It the Jansky VLA
By the early 2000s, the VLA’s electronics were decades old, and its sensitivity lagged behind what modern technology could achieve. Between 2001 and 2012, every major electronic system in the array was replaced in a project initially called the Expanded Very Large Array. The upgrade delivered roughly a tenfold improvement in continuum sensitivity across frequencies from 1 to 50 gigahertz and increased the instantaneous bandwidth to as much as 8 gigahertz in both polarizations.1arXiv. Expanded Very Large Array The heart of the overhaul was a new correlator called WIDAR, which can process up to 16 gigahertz of total bandwidth and divide each antenna-pair signal into at least 16,000 individual frequency channels.2arXiv. Expanded Very Large Array That spectral flexibility is critical for studies of gas in galaxies and other work where astronomers need to separate signals at closely spaced frequencies.
The practical effect was enormous. Observations that once required hours of telescope time to reach a given depth could now be done in minutes, freeing the array for a much broader range of projects. Surveys that would have been impractical with the old electronics became routine, and the array could respond faster to unexpected events like gamma-ray bursts and gravitational-wave alerts.
Radar Imaging of Mercury’s Polar Ice
One of the VLA’s more surprising contributions came in planetary science. In 1991, scientists used the 70-meter Deep Space Network antenna at Goldstone, California, to bounce a continuous radar beam off Mercury while the VLA received and mapped the reflected signals. The result was the first full-disk radar image of Mercury, covering roughly 77 percent of the planet’s surface at resolutions as fine as 150 kilometers. Near the north pole, the images revealed a patch of unusually bright radar reflectivity with polarization characteristics consistent with ice.3Journal of Geophysical Research: Planets. Mercury: full‐disk radar images and the detection and stability of ice at the North Pole
The idea that Mercury, the closest planet to the Sun, could harbor ice sounds counterintuitive. But Mercury’s spin axis is nearly perpendicular to its orbital plane, so the floors of deep craters near the poles never see sunlight. Temperatures there can remain far below freezing despite the scorching dayside. The VLA-Goldstone detection kicked off decades of follow-up work. Later radar campaigns at Arecibo mapped additional ice-bright features at both poles with finer resolution.4Icarus. Radar imagery of Mercury’s putative polar ice: 1999–2005 Arecibo results NASA’s MESSENGER spacecraft, which orbited Mercury from 2011 to 2015, confirmed the presence of both water ice and dark organic-rich material overlying the ice in permanently shadowed craters. Modeling of the dark deposits suggests they extend beyond the permanently shadowed zone and correspond to surface temperatures consistent with the stability of complex organic compounds rather than water ice alone.5The Planetary Science Journal. New Illumination and Temperature Constraints of Mercury’s Volatile Polar Deposits
Jets, Galaxies, and Magnetic Fields
Much of the VLA’s bread-and-butter work involves galaxies and the supermassive black holes at their centers. When material spirals into a black hole, twin jets of plasma can be launched at close to the speed of light, and these jets shine brightly at radio wavelengths. Detailed VLA observations of the galaxy NGC 4258 revealed a compact nuclear radio source with a subtly inverted spectrum stretching from centimeter wavelengths all the way to millimeter wavelengths. The jet properties, including the ratio of brightness between the approaching and receding sides, point to a mildly relativistic outflow viewed at a specific angle, which helps astronomers understand how jet physics varies from one galaxy to another.6The Astrophysical Journal. Nuclear Radio Jet from a Low-luminosity Active Galactic Nucleus in NGC 4258
The VLA is equally useful for mapping the quiet gas that fills galaxies between episodes of dramatic activity. Neutral hydrogen emits a faint radio line at 21 centimeters, and the VLA can map this emission across entire galaxy disks with enough detail to trace how gas moves in response to a galaxy’s gravitational field. Studies of the barred spiral NGC 3319, for instance, used the VLA to chart the distribution and motion of neutral hydrogen across the galaxy at spatial resolutions between 11 and 50 arcseconds, revealing how the bar-shaped structure channels gas inward.7Monthly Notices of the Royal Astronomical Society. A neutral hydrogen study of the barred spiral galaxy NGC 3319
On still larger scales, the VLA has been used to probe the magnetic fields threaded through galaxy clusters. By measuring the Faraday rotation of polarized radio signals passing through the Coma cluster, researchers mapped magnetic field structures at kiloparsec resolution across one of the nearest rich clusters of galaxies.8Astronomy & Astrophysics. The Coma cluster magnetic field from Faraday rotation measures Faraday rotation occurs when polarized light passes through magnetized plasma; the amount of rotation tells you the strength and direction of the intervening magnetic field. Cluster-scale magnetic fields are still poorly understood, and observations like these provide some of the only direct measurements available.
Watching Planet Formation in Action
Young stars are often surrounded by rotating disks of gas and dust, and planets form inside these disks. The VLA operates at wavelengths well suited to detecting the thermal glow of centimeter-sized dust grains, which are an intermediate stage between microscopic interstellar particles and fully formed rocky bodies. This sensitivity to larger grains complements shorter-wavelength observatories that are more sensitive to smaller dust.
A landmark study of TW Hydrae, a roughly 10-million-year-old star only about 190 light-years away, used VLA observations at 7 millimeters alongside other data to build a detailed model of the star’s disk. The model required dust grains that had grown to about a centimeter in size in the outer disk, while the inner region within about 4 astronomical units was mostly evacuated, with only a trace of fine particles remaining. The researchers interpreted the sharp transition between the cleared inner zone and the dense outer disk as a possible signpost of a growing planet carving a gap in the material around it.9The Astrophysical Journal. Evidence for a Developing Gap in a 10 Myr Old Protoplanetary Disk
The proposed next-generation VLA would push this science much further. Simulations show that observations at 3 millimeters with angular resolution of about 5 milliarcseconds could reveal disk gaps and asymmetries caused by planets with masses as low as a few times that of Earth in the nearest star-forming regions, a threshold far below what current instruments can reach.10The Astrophysical Journal. Investigating the Early Evolution of Planetary Systems with ALMA and the Next Generation Very Large Array
Fast Radio Bursts and Gravitational-Wave Counterparts
The upgraded VLA has become a key player in time-domain astronomy, the study of objects that change or appear suddenly. Fast radio bursts are one of the hottest topics in the field: millisecond-duration flashes of radio energy arriving from cosmological distances, whose origins are still being worked out. Because the VLA can localize sources to a fraction of an arcsecond, it can pin a fast radio burst to a specific host galaxy, which is essential for understanding what produces them. The VLA’s “realfast” search system continuously sifts through the array’s data stream looking for these brief signals and has successfully discovered and localized new bursts.11The Astrophysical Journal. A Distant Fast Radio Burst Associated with Its Host Galaxy by the Very Large Array
Gravitational-wave astronomy has opened another frontier where the VLA contributes. When the LIGO and Virgo detectors pick up ripples in spacetime from merging compact objects, the gravitational-wave signal alone gives only a rough patch of sky where the event occurred, sometimes spanning tens or even hundreds of square degrees. Radio telescopes can search that patch for an electromagnetic counterpart. After the binary black hole merger GW151226, the VLA imaged a 100-square-degree error region to test rapid wide-field search techniques, demonstrating the feasibility of blind radio follow-up even when no bright counterpart was expected.12The Astrophysical Journal. A Case Study of On-the-fly Wide-field Radio Imaging Applied to the Gravitational Wave Event GW151226
The payoff came in 2017, when LIGO and Virgo detected GW170817, the first confirmed merger of two neutron stars. Unlike black hole mergers, neutron star collisions produce matter-rich outflows that interact with surrounding gas and generate radio emission. The VLA began observing the merger’s location within hours, obtaining the earliest centimeter-band data at just under 14 hours after the event. Faint emission at 6 gigahertz appeared about 19 days later and strengthened over subsequent weeks, tracing the expanding blast wave as it plowed into surrounding material.13The Astrophysical Journal Letters. The Electromagnetic Counterpart of the Binary Neutron Star Merger LIGO/Virgo GW170817. VI. Radio Constraints on a Relativistic Jet and Predictions for Late-time Emission from the Kilonova Ejecta This kind of late-time radio monitoring is especially valuable because theoretical models predict radio afterglows peaking hundreds of days after a merger, a timescale well suited to the VLA’s scheduling flexibility.14The Astrophysical Journal. A Case Study of On-the-fly Wide-field Radio Imaging Applied to the Gravitational Wave Event GW151226
Surveying the Entire Radio Sky
The VLA Sky Survey, known as VLASS, is an ongoing effort to image every part of the sky visible from New Mexico, everything above a declination of negative 40 degrees, at a resolution of about 2.5 arcseconds and frequencies between 2 and 4 gigahertz with full polarization information.15arXiv. The Karl G. Jansky Very Large Array Sky Survey (VLASS). Data Products The survey is designed to be repeated, so that comparing images from different epochs reveals objects that have brightened, faded, or moved. That makes VLASS a discovery engine for transient and variable sources, from stellar flares to tidal disruption events to new fast radio burst hosts.
Processing the data is a serious computational challenge. Quick-look images in total intensity are produced within weeks of observation, but higher-accuracy images that exploit the full dataset require corrections for the curvature of the sky, which can increase the computational cost per image by a factor of roughly 100 for half the survey area.16arXiv. The Karl G. Jansky Very Large Array Sky Survey (VLASS). Data Products The resulting data products are publicly available, so researchers worldwide can mine them without needing their own telescope time.
Searching for Extraterrestrial Signals
The VLA is not just for conventional astrophysics. A system called COSMIC, for Commensal Open-Source Multimode Interferometer Cluster, has been installed on the array to search for technosignatures, artificially produced radio signals that might indicate the presence of extraterrestrial intelligence. COSMIC operates commensally, meaning it piggybacks on whatever observations other astronomers have already scheduled, copying the data stream and analyzing it in parallel without interfering with the primary science program.17The Astronomical Journal. COSMIC: An Ethernet-based Commensal, Multimode Digital Backend on the Karl G. Jansky Very Large Array for the Search for Extraterrestrial Intelligence
The advantage of running a technosignature search on the VLA, rather than a single dish, is resolution. A single dish can detect a suspicious signal but cannot pinpoint where on the sky it came from with much precision. The VLA’s interferometric baseline can localize a candidate signal to a tiny patch of sky, making it far easier to rule out terrestrial interference or identify a specific star system as the source. Because COSMIC rides along with all VLA observations, it accumulates an enormous volume of sky coverage over time without competing for dedicated observing hours.
The Next-Generation Very Large Array
Planning is well underway for a successor facility called the next-generation Very Large Array, or ngVLA. The reference design calls for roughly 244 antennas of 18 meters in diameter and an additional 19 antennas of 6 meters, operating across a frequency range from 1.2 to 116 gigahertz.18arXiv. Science with an ngVLA: The ngVLA Reference Design The antennas would be distributed across a much larger footprint than the current VLA, with a dense core in New Mexico and long-baseline stations extending across North America. That combination would deliver both the surface-brightness sensitivity needed for diffuse emission and the angular resolution needed to zoom in on compact sources.
The science case for the ngVLA covers territory the current VLA can only scratch the surface of. At 3 millimeters, the ngVLA could image planet-forming gaps in disks caused by bodies just a few times Earth’s mass in nearby star-forming regions.19The Astrophysical Journal. Investigating the Early Evolution of Planetary Systems with ALMA and the Next Generation Very Large Array At centimeter wavelengths, it could detect thermal emission from the surfaces of nearby rocky exoplanets, track the chemical evolution of galaxies at high redshift, and monitor thousands of transient events simultaneously. The jump in collecting area from 27 antennas to more than 240 would push the sensitivity roughly an order of magnitude beyond even the upgraded Jansky VLA, opening observational parameter space that no existing facility can access.
Construction timelines for large observatories tend to stretch, and the ngVLA is still in the design and review phase. But the current VLA’s scientific output, more than four decades after its dedication, is a powerful argument that a well-designed general-purpose radio array can remain productive and adaptable far longer than any individual research program. The 27 dishes on the Plains of San Agustin have been quietly central to discoveries their original designers never imagined, from polar ice on Mercury to the radio glow of merging neutron stars. Whatever form its successor takes, the VLA has set the template for what a radio observatory can be.