From Cold War Spy Satellite to Cosmic Observatory: NASA's Roman Telescope

In 2012, Alan Dressler walked into a clean room at a defense contractor in Rochester, New York, and found himself staring at a row of pristine mirrors—each one a near-perfect replica of the Hubble Space Telescope. They weren't built for astronomy. They were spy satellites, engineered in the 1990s and early 2000s by the National Reconnaissance Office to peer down at Earth with extraordinary precision, capable of spotting objects smaller than a coffee mug from orbit. One of these telescopes, born from a classified military program that had cost $13 billion more than planned before being scrapped in 2005, was about to be given to NASA for free. It would become the Nancy Grace Roman Space Telescope, now sitting in Florida awaiting launch on a SpaceX Falcon Heavy rocket, ready to peer not down at Earth but up at the cosmos—hunting for the invisible forces that hold the universe together.

The story of Roman's existence is one of bureaucratic accident and scientific fortune. In 2010, the National Academies of Sciences told NASA to build a telescope called WFIRST to study dark energy and dark matter, the mysterious components that make up more than 95 percent of the universe. The project faced immediate skepticism. Astronomers wanted something more ambitious, something that could also hunt for exoplanets. Congress was pouring money into the James Webb Space Telescope, which had already ballooned to $10 billion. WFIRST looked doomed before it even began. Then, at an American Astronomical Society meeting in 2011, theoretical astrophysicist David Spergel pulled Alan Dressler aside with an extraordinary whisper: the National Reconnaissance Office had spare telescopes it no longer needed. Would NASA want them?

The offer came from a failed military program called Future Imagery Architecture. After the September 11 attacks, the NRO had doubled down on building a new generation of spy satellites with Hubble-sized mirrors. Contracted to Boeing, the program spiraled—costs ballooned, schedules slipped. By 2005, officials decided to kill it, but not before hardware had already been constructed and sat gathering dust in a clean room. Michael Moore, then NASA's acting deputy director for astrophysics, had heard whispers about the surplus equipment through his Air Force contacts. He asked about it in the mid-2000s and was told no. By 2011, the answer had changed. The NRO offered NASA two partially disassembled telescopes with 2.4-meter mirrors—the same size as Hubble's—plus a third primary mirror and spare components. An NRO representative traveled to NASA headquarters in a secure room and made the formal offer to John Grunsfeld, the agency's top science official.

When Dressler, Grunsfeld, and other scientists traveled to Rochester to see the telescopes in person, they found themselves standing before hardware that would transform the mission. The spy telescope's 2.4-meter mirror was twice the size of what WFIRST had originally been designed to use. More importantly, the larger mirror made room for a coronagraph—an instrument that could block starlight and reveal orbiting planets, the very capability that exoplanet researchers had been demanding. NASA administrator Charlie Bolden approved the gift. It would still take more than a decade and roughly $4.3 billion to strip away the classified components, build new instruments, and engineer the transformation from spy satellite to space observatory. L3Harris, which had acquired the defense contractor Exelis, held the mirror in Rochester and managed the conversion. The fate of the other telescopes remains unknown; NASA, the NRO, and L3Harris have not disclosed whether they still exist or where they are stored.

Even after the gift arrived, Roman nearly died repeatedly. The mission was canceled five times in presidential budgets—twice under Obama, three times under Trump. Each time, Congress voted to save it, with astronomers including Spergel traveling to Washington to lobby senators like Chuck Schumer of New York. To make the mission permanent, Thomas Zurbuchen, then head of NASA's science projects, decided to name it. Naming a mission, he reasoned, makes it politically unshakeable. He chose Nancy Grace Roman, the astronomer who had died in 2018 at 93 and had been NASA's first chief of astronomy in the 1960s, a rare woman in the field who had championed the Hubble Space Telescope so effectively that she earned the nickname "Mother of Hubble." It was the first space telescope named for a woman.

When Roman launches to the second sun-Earth Lagrange point—one million miles from Earth, where the James Webb Space Telescope also orbits—it will carry a field of view at least 100 times larger than Hubble's and a 300-megapixel camera so powerful that displaying a single image would require a wall of 4K televisions. In five years of observations, it will survey about 12 percent of the sky and image billions of galaxies. It will search for the subtle warping of light caused by dark matter bending spacetime—a technique called weak gravitational lensing that will map how matter is distributed across the universe and how that distribution has changed over cosmic time. It will hunt for thousands of type Ia supernovae stretching back more than 10 billion years, revealing how the universe's expansion has accelerated or decelerated across history. It will measure baryon acoustic oscillations, the frozen echoes of sound waves that traveled through the early universe, using the distance between galaxies as a cosmic ruler to measure dark energy's strength.

Roman will also transform exoplanet science. Using a technique called microlensing—detecting the gravitational tug of planets bending the light of distant stars—it should discover roughly two thousand planets in the dense stellar population of the galactic bulge, including worlds as small as Earth's moon. Using the transit method, watching for the dips in starlight as planets cross their stars, it should find perhaps 100,000 additional planets, revealing multiple times more worlds than have been discovered in all of human history. But the coronagraph may be Roman's most consequential instrument. It will suppress starlight to a ratio of one part in a billion, allowing it to image planets the size of Jupiter by their reflected light. If successful, it will be the first time humanity has directly seen a cooler exoplanet reflecting its star's light rather than glowing from its own heat. It might even reveal ring systems around distant worlds. This coronagraph is a test bed for technology that NASA plans to fly on the Habitable Worlds Observatory in the 2040s, a telescope designed to image 25 Earth-like worlds and analyze their atmospheres for signs of life. The first detection of life beyond our solar system, if it comes, may owe everything to a collapsed spy program, a bureaucratic offer, and a series of near-miraculous saves. "All I know," says Marc Postman, head of the Science Mission Office at the Space Telescope Science Institute, "is that we got a good mirror."