China has postponed the launch of its Xuntian Space Telescope amid an international race to chart the frontiers of modern cosmology.

Now slated for liftoff from southern China’s Wenchang Space Launch Center in mid-2025, the two-meter Xuntian ( “Survey the Heavens”) will join the European Space Agency’s 1.2-meter Euclid space telescope, which released its first full-color images this month, and NASA’s 2.4-meter Nancy Grace Roman Space Telescope, set for a mid-2027 launch, to survey billions of distant galaxies, map the structure of the universe and test contending theories of dark matter and dark energy.

“Xuntian was planned to launch by the end of this year. The time line is now adjusted to June 2025,” says Zhan Hu, project scientist of Xuntian’s telescope system at the National Astronomical Observatories of China in Beijing. Zhan and his team are now finishing their work on a preflight “engineering qualification model” for Xuntian that will begin rigorous performance tests early next year, he says.

In a challenging first, China is domestically developing all five instruments on Xuntian, Zhan says. He leads a team of about 100 engineers and scientists from five research institutes across the country that is working on a 2.6-gigapixel survey camera that will be the telescope’s main instrument.

Xuntian is one of the most important scientific facilities China has ever built, says Quentin Parker, an astrophysicist at the University of Hong Kong—and that makes its delay surprising. “It’s unusual for China because they don’t normally put things off. They’ve been excellent at keeping their missions on track,” he says.

The delay could have major implications for the three-way race to solve the dual mysteries of dark matter—the invisible gravitational glue that allows galaxies to form—and dark energy—the mysterious but dominant force behind our universe’s ever accelerating expansion. Together dark matter and dark energy constitute an overwhelming 95 percent of the universe’s mass and energy, with familiar matter making up the 5 percent remainder. Learning the true nature of dark matter and dark energy is critical to cosmology, potentially offering answers to questions concerning the universe’s deepest origins, eventual fate and most everything in between. A later launch for Xuntian reduces the temporal edge it might otherwise have over NASA’s Roman telescope, Parker says. “One great advantage of launching before your competitor is that you get the first bite of the cherry for the science,” he adds. If the launch dates of Xuntian and Roman turn out to be close, there will “be an interesting dynamic in terms of who will get the first data, first images and first research results.”

After decades of waiting, China’s astronomers are understandably eager to have their own observatory that is comparable to the Hubble Space Telescope, says astrophysicist Wu Xuebing of Peking University in China. Given Xuntian’s state-of-the-art design and cutting-edge technologies, however, “delay is not necessarily a bad thing. It’s important to make sure that everything works before it goes into space,” Wu says.

Xuntian is indeed ambitious. First approved in 2013 as part of China’s plans for a space station, the mission’s concept and design has evolved over time to boast, among other things, a truly panoramic field of view that is more than 300 times larger than Hubble’s. This means Xuntian—which is also sometimes called the Chinese Space Station Telescope—can, with a single snapshot, survey a swath of sky that would take Hubble almost a year to image and do so with roughly the same resolution. During each observation, Xuntian also sees twice as much sky as Euclid and four times as much as Roman.

Its survey camera, equipped with charge-coupled device detectors that are packed with 2.6 billion pixels, aims to cover 17,500 square degrees—or 40 percent—of the entire sky during its planned decade-long operation some 400 kilometers above the ground in the same orbit as China’s space station Tiangong.

Observing in the near-ultraviolet and optical wavelengths between 0.255 and one micron, Xuntian will be “perfectly complementary” to Euclid and Roman, which focus more on the near-infrared, says Yun Wang, a cosmologist at the Infrared Processing and Analysis Center at the California Institute of Technology.

All three missions share a common methodological cornerstone, charting the distances and distributions of galaxies to derive deeper cosmic measurements. But each will sample the universe at different ages, albeit with some overlap. Xuntian will look back in time to galaxies aglow when the universe was one third of its current age. Meanwhile Euclid and Roman will focus on galaxies from halfway to three quarters of the way back through the universe’s nearly 14 billion years of history.

Even when different telescopes are measuring exactly the same thing, it’s still crucial to cross-check their results, says Jason Rhodes, an astrophysicist at NASA’s Jet Propulsion Laboratory, who works on both Euclid and the forthcoming Roman. “The effects dark energy has on things we can observe are small on each individual galaxy,” so the measurements are very demanding—and even minuscule instrumental errors can yield wildly wrong results, he says.

Xuntian, Euclid and Roman will all also use an observational technique called weak gravitational lensing to map dark matter by spotting tiny distortions in the shapes of galaxies. Such distortions are caused by clumps of intervening dark matter that, via their spacetime-warping gravitational fields, subtly alter the path of light from galaxies as it speeds toward Earth. Unlike strong lensing, in which a massive foreground galaxy can stretch light from a dotlike background galaxy to look like a curve, weak lensing only twists the images of galaxies by a thousandth or less of the ellipticity of their apparent angular size and is extremely challenging to measure.

According to Zhan, Xuntian’s optical system has an edge in this regard because its secondary mirror will be placed off to the side rather than directly in front of the primary mirror to avoid blocking any incoming light and creating diffraction patterns in the images. This so-called off-axis design distinguishes Xuntian from Hubble, Euclid and Roman, all of which use an on-axis architecture that invariably projects diffraction “spikes” and other visual artifacts onto resulting images. Xuntian’s spike-free images will consequently help reduce errors in weak lensing analysis, Zhan says.

While Xuntian will spend the vast majority of its time chasing dark matter and dark energy by surveying far-distant galaxies, it also has a long list of secondary science goals to fulfill with the same survey data and the observations of four other instruments. For instance, it will search for exoplanets around a sample of nearby stars using a starlight-blocking coronagraph that can allow a star’s far fainter accompanying planets to be seen. Additionally, the telescope will include a high-sensitivity terahertz receiver to study the chemistry of giant molecular clouds and other complex objects in the Milky Way and neighboring galaxies; a multichannel imager to carry out more focused extremely deep-field observations and to monitor rapidly changing phenomena such as tumbling asteroids and detonating supernovae; and an integral field spectrograph to probe the extreme physics of matter swirling around and into black holes.

And after all five instruments are assembled onto the telescope platform, there will be an empty slot for later use by a domestic, foreign or jointly developed device installed by astronauts from the Tiangong space station, Zhan says.

Across Xuntian’s first decade of operations, several rendezvous and docking maneuvers are planned between the telescope and the space station in low-Earth orbit to allow for refueling, maintenance and upgrades. Learning from Hubble, Xuntian’s architects considered such servicing crucial for ensuring the observatory’s enduring scientific competitiveness.

“There’re a lot of things to like about co-orbiting,” says Jonathan McDowell, an astronomer at the Center for Astrophysics | Harvard & Smithsonian. For instance, fixing or swapping instruments on Euclid or Roman would be prohibitively difficult because these two telescopes are each stationed at Lagrange point 2, a position between the sun and Earth that is about 1.5 million kilometers away from our planet.

But Xuntian’s perch in low-Earth orbit also means that our looming planet will constantly block its view of nearly half of the sky, limiting the telescope’s observing efficiency, McDowell says. Additionally, Xuntian’s orbit will cause the telescope to transition between day and night every 90 minutes or so, creating thermal instabilities that can affect its instruments, Rhodes points out.

Such issues, however, may prove to be the least of Xuntian’s orbital concerns. “My biggest worry for Xuntian is that since it has a large, wide field of view, and since it’s below [SpaceX’s] Starlink satellites, it’s going to see an awful, awful lot of Starlink satellite trails across all of its images,” McDowell says.

Zhan’s team has used simulations to estimate the negative impacts that Starlink and other satellite constellations will have on Xuntian. Once the 40,000-plus Starlink satellites and similar soon-to-debut projects are fully operational in orbit, Zahn says, Xuntian’s survey camera will usually see at least one satellite in each of its main camera’s 150-second exposures. “But they appear to be relatively easy to identify and take out of the data,” he adds.

So far, the Chinese astronomy community has received funding to establish four centers that will coordinate and support Xuntian’s research once the telescope is operational. There are also grants earmarked for Xuntian-related preparatory studies, including simulations of the telescope’s imaging, operations and data processing. Such in-depth, far-reaching support efforts are unprecedented for space missions in China.

“The Xuntian team has gotten technical ambitions, and they’ve come this far,” Wang says. With Xuntian, Euclid and Roman sharing observational data or even coordinating their surveys, scientists will hopefully soon place stronger constraints on dark energy theories. “We probably won’t have the ultimate answer, but they will give us clues to carry on. I’d say we can expect breakthroughs within 10 years,” she says.