NASA’s Roman Space Telescope Is Go for a Wide-Field Launch

NASA and SpaceX polled the Nancy Grace Roman Space Telescope go for launch after Friday’s review, targeting 7:26 a.m. EDT Sunday from Kennedy Space Center. Overnight into Saturday, Falcon Heavy rolled out of the hangar at Launch Complex 39A, the step that puts the stacked rocket on the pad.

The observatory is flying about eight months earlier than NASA’s formal May 2027 readiness date. Weather, not the hardware poll, is the live risk, with the U.S. Space Force calling a 60% chance of favorable conditions.

NASA and SpaceX Polled Go After Friday’s Review

The Launch Readiness Review at Kennedy on Friday, Aug. 28, is the last major gate before countdown. NASA’s Roman blog said the poll covered Falcon Heavy, the observatory, weather, and the teams that will staff the attempt. A Flight Readiness Review on Aug. 21 had already cleared the spacecraft to be sealed in its fairing.

NASA Kennedy posted the result in plain language: mission teams polled “GO.”

.@NASARoman is GO for Launch!

NASA and SpaceX mission teams polled “GO” during today's Launch Readiness Review.

Liftoff of NASA's Nancy Grace Roman Space Telescope is targeted for 7:26 a.m. EDT Sunday Aug. 30 on a SpaceX Falcon Heavy from Kennedy! https://t.co/qRgUn7pLGO pic.twitter.com/YPQ1OdawIg

— NASA's Kennedy Space Center (@NASAKennedy) August 28, 2026

Sunday is the sixth NASA primary payload SpaceX has flown from Kennedy on a Falcon, after IXPE, Psyche, GOES-U, Europa Clipper, and IMAP. Coverage starts at 6:20 a.m. EDT. If Sunday slips, SpaceX is holding Monday, Aug. 31, at 7:22 a.m. ET. NASA scheduled a science briefing for 9 a.m. Saturday and a prelaunch news conference at 10:30 a.m., with a postlaunch briefing set for 9:30 a.m. Sunday.

THE SUNDAY ATTEMPT

• Targeted time: 7:26 a.m. EDT Sunday, Aug. 30, from Launch Complex 39A.
• Weather odds: Space Launch Delta 45 meteorologists give a 60% chance of favorable conditions, with cumulus clouds and surface electric fields as the main constraints.
• Backup window: 7:22 a.m. ET Monday, Aug. 31, if Sunday is waved off.
• Where to watch: NASA coverage begins at 6:20 a.m. EDT on NASA Live, with SpaceX starting a webcast about an hour before liftoff.

SpaceX lists a 7:10 to 7:30 a.m. ET window around that 7:26 target. The 45th Weather Squadron’s Friday forecast put Sunday’s pad temperature near 80°F with 90% humidity and 10 to 15 mph winds, and it flagged Monday as slightly cleaner on constraints. Launch commentators watching the hangar said the three-core stack began rolling in the early hours of Saturday, which matches the usual last roll to 39A before a morning attempt.

The Same 2.4-Meter Mirror, a Hundred Times More Sky

Roman uses a 2.4-meter primary, the same width as Hubble’s, and NASA built it to do the job Hubble never had time for: map huge patches of sky at Hubble-like sharpness. Newer glass and a three-mirror layout cut the primary to 410 pounds (186 kilograms), against Hubble’s 1,825 pounds (828 kilograms). NASA says the shorter focal length, about f/7.9 at the wide-field camera, trades Hubble’s long zoom for a view at least 100 times larger per pointing.

The Space Telescope Science Institute gives the Wide Field Instrument a field of view of 0.281 square degrees on 18 H4RG-10 detectors, about 300 million pixels, and calls that patch roughly 100 times the area of Hubble’s ACS or Webb’s NIRCam and 200 times Hubble’s WFC3 infrared camera. NASA’s Hubble comparison page says that in more than 30 years Hubble has seen one-tenth of one percent of the sky, and that in five years Roman will image 50 times as much sky as Hubble captured in 30 years.

HUBBLE AND ROMAN SIDE BY SIDE

STScI’s instrument guide says the camera can cover Andromeda’s main body in a few pointings and survey that galaxy nearly 1,500 times faster than Hubble. NASA’s 2020 development approval used a similar yardstick: Hubble needed more than 650 hours to mosaic pieces of Andromeda, while the wide-field design could cover the whole galaxy in about three hours. Hubble remains the better ultraviolet zoom. Roman is the infrared wide-angle lens.

Think of Roman as a wide-angle lens in infrared light while Hubble is the zoom lens in ultraviolet and visible light.

NASA Science, Hubble vs. Roman

Julie McEnery, Roman’s senior project scientist at Goddard, told a July briefing the surveys will touch every area of astronomy, SpaceNews reported. That is the sleeper hiding under a launch-time headline: Webb stares deep, Hubble inspects, and Roman is the one that will write the catalog they both use.

Why Falcon Heavy Is Flying Roman to L2

Roman needs a high-energy ride to Sun-Earth Lagrange Point 2, the same neighborhood as Webb, about a million miles from Earth. Falcon Heavy is flying that path from pad 39A, and SpaceX is recovering only the side boosters. NASA picked the vehicle under a Launch Services contract awarded in July 2022. After fairing separation and a second upper-stage burn, Roman deploys at about T+31 minutes 31 seconds, per SpaceX’s timeline.

THE LAST TWO WEEKS AT KENNEDY

1. August 21, 2026: Teams encapsulate Roman in the payload fairing at the Payload Hazardous Servicing Facility and complete the Flight Readiness Review.
2. August 25, 2026: The encapsulated observatory moves to SpaceX’s hangar at Launch Complex 39A for mating to Falcon Heavy.
3. August 28, 2026: NASA and SpaceX complete the Launch Readiness Review and poll the mission go for Sunday.
4. August 29, 2026: Falcon Heavy rolls from the hangar toward the 39A pad ahead of the 7:26 a.m. attempt.

SpaceX says this is the first flight for one side booster and the third for the other, after the GOES-U and Viasat-3 F3 missions, and that the side boosters land at LZ-2 and LZ-40 at Cape Canaveral Space Force Station about 7 minutes 40 seconds after liftoff. The company does not list a landing for the center core, which is the usual cost of the energy this trajectory needs. That is why the stack can look uneven on the pad: the side cores are on different flight lives, and the middle one is not coming home.

Jackie Townsend, Roman’s project manager at Goddard, told reporters in July the project was where it needed to be as launch approached, SpaceNews reported, after more than 50 mission-readiness tests and 1,100 hours of console time for the first 100 days. The cruise to L2 takes about three months of deployments, calibrations, and tests before science operations start.

Named for the Astronomer Who Fought for Hubble

NASA renamed the mission on May 20, 2020, dropping the WFIRST acronym for Nancy Grace Roman (1925-2018), NASA’s first chief of astronomy and the official NASA still calls the “mother” of Hubble. She pushed space telescopes that would work above the atmosphere and share their data widely, and she helped drive the Great Observatories that followed Hubble, including Chandra and the retired Compton and Spitzer missions. Then-administrator Jim Bridenstine said in NASA’s naming release he could think of no better name for the spacecraft then still called WFIRST, which he described as the successor to Hubble and Webb.

The hardware carries a quieter tribute. NASA says more than 1.3 million people submitted names this summer for a memory card on a plaque aboard the spacecraft, and those names ride to L2 with it.

The program itself almost did not. NASA held WFIRST to a cost cap and, in 2017, recast the coronagraph as a technology demonstration to keep the bill in check. The first Trump administration proposed zeroing the mission in several budget years; Congress put the money back. A June 2021 replan, tied to COVID delays, set a life-cycle cost of $4.3 billion and the May 2027 launch-readiness date, a figure the Government Accountability Office later treated as the agency’s commitment. Townsend told Ars Technica the team operated to that cap. SpaceNews, reporting this week, put the same $4.3 billion on the five-year prime mission now sitting on the rocket.

The 2.4-meter blank itself came from a National Reconnaissance Office donation, a fact NASA and several science outlets have long noted, which is why a Hubble-sized mirror could be flown without grinding a new one from scratch. For a flagship, the unusual line is the one NASA keeps repeating: it is launching early, and it stayed inside the cap Webb never saw.

Surveys That Will Count Galaxies by the Billion

Roman’s prime mission is five years, with a 10-year goal, from L2. NASA designed the spacecraft to be refueled, even though the agency has no servicing craft at L2 today. Fuel is expected to set the lifetime. Core Community Surveys, shaped with input NASA said came from more than 1,000 scientists at 350-plus institutions, will take most of the time. The data become public as soon as they are processed.

THE CORE SKY PROGRAMS

• High-latitude map: The High-Latitude Wide-Area Survey images about 2,400 square degrees in several bands plus grism spectra, then adds about 2,700 square degrees more in H-band, for more than 5,000 square degrees NASA will use to weigh dark energy and dark matter through weak lensing, galaxy clustering, and supernovae.
• Galactic bulge watch: A time-domain stare at the Milky Way’s crowded heart monitors on the order of 100 million stars for microlensing, the brief brightening when a foreground mass crosses a background star, which NASA expects will turn up about 2,500 planets, including worlds in wide orbits and free-floating planets Kepler-style transits miss.
• Open follow-up: The same wide frames will feed exploding stars, growing black holes, outer solar system objects, and guest programs; ESA also expects more than 100,000 transiting planets as a byproduct of the bulge monitoring.

ESA’s Roman factsheet puts Euclid’s larger single-pointing field of view at 0.53 square degrees against Roman’s 0.28, while Roman flies the larger 2.4-meter mirror and sharper infrared images. NASA says the overlap lets scientists use Roman’s higher-quality patches to calibrate Euclid’s much wider map, then pair both with Vera C. Rubin Observatory’s ground-based optical survey. ESA also estimates Roman will gather about 20 petabytes over the five-year prime mission. A Nature note this week said exoplanet researchers are watching because a successful microlensing census could multiply the known planet list many times over; NASA’s own planning figure remains those roughly 2,500 microlensing worlds, plus the transit haul.

Webb will still take the deep portraits. Roman’s job is the map that tells Webb, Hubble, and Chandra where to look next, and it will do that while Euclid and Rubin are already sweeping the same sky in other bands.

The Coronagraph Flies as a Test for Earth Twins

Tucked beside the survey camera is the instrument NASA has spent years trying not to oversell. The Coronagraph Instrument, built at the Jet Propulsion Laboratory, is a technology demonstration: masks, sensors, and deformable mirrors that flex to cancel starlight so a faint planet can be photographed in reflected light. NASA’s Hubble comparison page says the design should deliver 100 to 1,000 times the capability of earlier coronagraphs. STScI’s technical overview sets a post-processed contrast goal better than 10 to the minus 8. Scientific American described a pair of flexible mirrors, each tied to 2,304 tiny pistons, as the active heart of the system.

NASA has said this is the first coronagraph in space to fly active wavefront sensing and control. It will not hunt Earth twins on this mission. It is meant to photograph older, colder giant planets in closer orbits than the hot young super-Jupiters that ground imagers already see, and to prove the plumbing a later Habitable Worlds Observatory would need to photograph an Earth-like planet. STScI’s overview gives the coronagraph a three-month observing allocation in the first 18 months. Vanessa Bailey, the coronagraph scientist at JPL, was on NASA’s Saturday science-briefing slate with McEnery and Shawn Domagal-Goldman, NASA’s astrophysics director.

If Sunday’s window holds, Roman reaches L2 after that three-month cruise and starts the checkout that NASA says should yield first science images by early 2027. The namesake spent her career arguing that NASA should put telescopes above the air and give the plates to everyone. The spacecraft on pad 39A is the survey version of that argument, finally stacked, eight months early, with a weather call still to make before 7:26 a.m.