NewSpace RFP

The Nancy Grace Roman Space Telescope launched on a SpaceX Falcon Heavy from Kennedy Space Center on August 30 at 11:26 UTC. Our launch record marks the flight successful. The mission, its cost reported at $4.3 billion in ESA's launch preview, carries the same 2.4-meter aperture as Hubble. It flew about nine months ahead of the agency's committed May 2027 launch date, and NASA calls the delivery on budget. What is new is not the mirror. It is what sits behind it: a 300.8-megapixel infrared camera whose single exposure covers 0.281 square degrees of sky, more than two hundred times the field of Hubble's infrared camera at the same sharpness.

Width as an instrument

Telescope design has a quantity for this: etendue, the product of collecting area and field of view, usually written in square meters times square degrees. Depth tells you how faint an object a telescope can reach; etendue tells you, to first order, how fast it can cover sky at that depth. For the flagship observatories the trade usually went to depth. Hubble's deep fields and Webb's first-year images are pencil beams, exquisite and narrow. Both observatories spend most of their time pointed at chosen targets. A survey telescope inverts the priorities. When the questions are statistical, and the biggest open questions in cosmology are, the instrument that wins is the one that can measure the shapes, distances and brightnesses of billions of objects, not hundreds.

Dark energy is the clearest case. Measuring whether cosmic acceleration has changed over time takes three measurements at scale: weak gravitational lensing across wide swaths of galaxies, supernovae in the thousands, and galaxy clustering over enormous volumes. No pointed telescope can visit that object by object. Exoplanet demographics is another. Roman's planned bulge survey monitors hundreds of millions of stars at high cadence for microlensing events. Microlensing finds planets near and beyond their stars' snow lines, in orbits pointed surveys mostly miss. Neither program is practical with a narrow field, however deep.

The namesakes

Both of the decade's big survey machines carry the names of women who shaped this branch of astronomy. Nancy Grace Roman was NASA's first chief of astronomy. Through the 1960s and 1970s she built the case, the coalition and the program office that produced the Hubble Space Telescope. The agency's own histories call her the mother of Hubble. She died in 2018, and NASA attached her name to this telescope in 2020. Vera Rubin, working with Kent Ford's spectrographs in the 1970s, showed that galaxy rotation curves stay flat where visible matter says they should fall. Those measurements moved dark matter from suspicion to mainstream problem. She died in 2016, and Congress named the observatory, until then the Large Synoptic Survey Telescope project, for her in 2019; the survey it will run is now the Legacy Survey of Space and Time. The pairing is apt. Rubin's observatory will find what changes in the sky; Roman's will weigh what the universe is made of. Both questions descend from measurements their namesakes made or made possible.

Why L2

Roman is headed where Webb already operates: a quasi-halo orbit around the second Sun-Earth Lagrange point, about 1.5 million kilometers anti-sunward of Earth. The rationale is thermal and geometric at once. At L2 the Sun, Earth and Moon all sit in one direction, so a single sunshade puts an infrared telescope in permanent shadow. Wide zones of sky away from that direction stay observable in long uninterrupted stretches. Hubble, in low Earth orbit, loses most targets to Earth's limb every 95 minutes and flexes as it crosses from sunlight to shadow. An infrared survey telescope in that orbit would spend its margin fighting its own temperature. The cost of L2 is that no servicing mission has ever flown there. Hubble was serviced by astronauts five times; Webb and Roman were engineered on the assumption that what launches is what you get.

Rubin answers the same requirements a third way: it stays on the ground, on Cerro Pachon in Chile, where an 8.4-meter monolithic mirror and a camera this large could never ride a fairing. The atmosphere is the price of aperture and serviceability. Its etendue of 319.5 square-meter-square-degrees, computed on its effective collecting area of roughly an unobscured 6.5-meter mirror, dwarfs anything in space. What the ground cannot do is Roman's territory: near-infrared imaging in bands where the atmosphere glows or absorbs, against a far darker sky, with a point-spread function that never sees turbulence.

The data

Width sets the pace, and the pace is severe. NASA rates Roman's survey speed at up to a thousand times Hubble's, the wide field and rapid slewing taken together. At that rate a month of Roman surveying covers more sky, at comparable depth, than Hubble could in 80 years. The downlink matches: about 1.4 terabytes of compressed science data per day, against roughly 3 gigabytes a day from Hubble and 50 to 60 from Webb. Over the five-year primary mission the archive approaches 20 petabytes.

All of it becomes public almost at once. Roman carries no proprietary period; observations land in the cloud-hosted Mikulski Archive shortly after they reach the ground, open to any researcher and anyone else. At these volumes the archive is the observatory: most Roman science will be done against the public dataset by people who never wrote an observing proposal.

The coronagraph

Roman's second instrument deserves more than the aside it usually gets. The Coronagraph Instrument is the first active coronagraph to fly in space. Behind its starlight-blocking masks sit two deformable mirrors, each about five centimeters across and backed by more than 2,000 actuators, reshaping themselves to cancel the flaws and drifts that scatter starlight into a planet's hiding place.

The performance target is planets about 100 million times fainter than the stars they orbit, up to a thousand times better than any space coronagraph before it. That is deep enough to see Jupiter-size planets by reflected starlight rather than their own residual heat, a first, and reflected light carries information about what an atmosphere is made of. The instrument is a technology demonstration. What it is demonstrating is the active wavefront-control toolkit a future mission will build on to reach Earth-size planets around Sun-like stars, which need roughly another hundredfold in contrast.

Fuel and servicing

NASA expects propellant to be Roman's life-limiting resource. The observatory was built for that problem: it is designed to be refueled in space, and its refueling interface was tested on the ground before launch. Refueling could extend operations well beyond the onboard propellant limit.

The catch is that no servicing mission of any kind has ever flown to L2. Hubble's five crewed servicing calls happened in low Earth orbit, a few hundred kilometers up; Roman sits 1.5 million kilometers out, beyond any crewed vehicle now flying. Extending Roman's life means a robotic refueling flight to a place nothing has ever been repaired, which makes the telescope a forcing function for an entire servicing capability.

How the four compare

The instrument suites tell the same story as the fields of view. Hubble carries four active instruments spanning imaging and ultraviolet spectroscopy, WFC3, ACS, STIS and COS, and remains the only one of the four that can see far-ultraviolet at all. Webb carries NIRCam, NIRSpec, MIRI and NIRISS, a full mid-infrared laboratory built for depth on individual targets. Rubin carries exactly one instrument, its camera, cycling its six filters across an 18,000-square-degree survey footprint every few nights. Roman sits between the philosophies. Its Wide Field Instrument does imaging and slitless spectroscopy through a grism and a prism across eighteen detectors. Its Coronagraph Instrument gets its own section above.

The complementarity is deliberate. Rubin finds and times; Roman confirms, weighs and reaches into the infrared where Rubin cannot; Webb inspects the individual objects the surveys surface; Hubble still owns the ultraviolet. The dark-energy answer, if this decade produces one, will be a joint custody arrangement.

Congratulations, Kevin, on your first launch.