Sunlight on Demand: The Fight Over Reflect Orbital’s Space Mirrors

On the evening of October 10, people in California’s Owens Valley looked toward Mount Whitney and saw something that did not belong there. A huge pale patch of light lay across the slopes below the highest peak in the lower 48 states, and it stayed lit from about 7 p.m. until midnight. The glow puzzled people as far away as Independence, more than 20 miles north.

It was a rehearsal. The light came from powerful spotlights on the ground, not from space. They were set up by Reflect Orbital, a US startup that wants to put giant mirrors into orbit and sell reflected sunlight to customers on the night side of the planet. Its own roadmap ends in 2035 with “50k+” satellites. That is the source of this week’s headlines about up to 50,000 sunlight-reflecting satellites, and of the anger of astronomers and environmental groups.

What Reflect Orbital is selling

Reflect Orbital calls itself “The Sunlight Company”, and its pitch is simple to say: “Noon light to moon light”, on demand, anywhere the satellites can reach. Each mirror would catch sunlight while it flies over the dark side of Earth and steer it down onto a patch of ground about 5 kilometers (3 miles) wide. Its list of uses runs from solar farms after sunset and disaster zones to farms, defense, events and even replacing streetlights.

The roadmap on its website climbs fast. Two satellites are listed for 2026, delivering about 0.1 lux for five minutes, which the company compares to a full moon. Then come dozens in 2027, more than 1,000 in 2028, more than 5,000 in 2030 and more than 50,000 in 2035, with up to 36,000 lux, which is daylight territory. Those are company targets. So far only one test satellite is licensed.

That satellite is called Earendil-1. The US Federal Communications Commission approved it on July 9, 2026, in an order from its Space Bureau. The license covers a two-year mission at about 625 kilometers up. Scientific American describes an 18-meter-wide steerable mirror, and recent reporting says the company now expects to launch next year. On October 14 it announced that it had bought two launches from the Indian company Skyroot Aerospace, on the Vikram-1 and Vikram-1U rockets. Chief executive Ben Nowack said there is “a real launch capacity shortage” and that the company wants to “earn the right to scale”.

The night they lit up Mount Whitney

The October test was meant to show what a satellite spot would look like on the ground, to the people the company sees as early customers: rescuers. “We mounted powerful mobile lighting near the peak of Mt. Whitney to recreate the light spot our satellites will deliver,” spokesperson Christopher Buscombe told SFGATE. Inyo County Search and Rescue and the China Lake Mountain Rescue Group ran drills under it.

Accounts of how bright it was differ. The San Francisco Chronicle and Sky & Telescope described the lit area as slightly brighter than a full moon. DarkSky International, which campaigns against light pollution, estimated it at about twice as bright as a full moon, spread over an area roughly 3 by 10 miles.

The rescuers were polite but not dazzled. Phillip Cash of China Lake Mountain Rescue told DarkSky that the light helped people see the terrain, but finding people was only “marginally better”. The glare hurt night vision: “Wow, I need a hat.” He also warned that a lost hiker who sees the landscape light up may start moving, which makes a search harder. The test targets were white, glossy boxes, not realistic stand-ins for people. And on cost, he said, “Most search and rescue teams… will not be able to afford it.” Sky & Telescope reported a planned price of $5,000 an hour.

The Bureau of Land Management told Sky & Telescope it “was not aware of or consulted” on the test, and one local, Jaime Lopez Wolters, called it “a whitewashing” of the product.

An old dream, and a Russian prototype

Mirrors in space are not a new idea. As the BBC has recounted, Hermann Oberth wrote about orbital reflectors in 1923, and Krafft Ehricke sketched a “Power Soletta” in the 1970s. The only real attempts came from Russia.

On February 4, 1993, a Progress cargo ship that had just left the Mir space station unfurled Znamya 2, a 20-meter mirror of thin reflective film. It cast a spot of light about 5 kilometers wide, roughly as bright as a full moon, that raced across Europe at 8 kilometers per second, from southern France toward western Russia. Clouds covered most of the route, and only a few people reported a flash. A larger Znamya 2.5 snagged on an antenna in February 1999 and was lost, and a 60- to 70-meter Znamya 3 was never built.

Reflect Orbital is betting that cheap launches and modern steering can do better. Its satellites would hold a spot on one target instead of sweeping across a continent, but Sky & Telescope reported that each could light a given place for only about four minutes. Continuous light means a lot of satellites.

How bright is “moonlight from space”?

In a 2026 paper in The Astrophysical Journal Letters, Miroslav Kocifaj of the Slovak Academy of Sciences, Gáspár Bakos of Princeton University and František Kundracik modeled what these mirrors would actually do. For Earendil-1’s 18-by-18-meter mirror at about 600 kilometers, they calculate a patch roughly 2.5 kilometers in radius lit to about 1.05 lux. That is around four times the 0.26 lux they use for a full moon. From inside the patch the mirror would shine at about magnitude -14.3, brighter than the full moon’s -12.7 (lower numbers are brighter).

Their numbers grow quickly with mirror size. A full-size 54-by-54-meter mirror reaches about magnitude -16.7 and lights the ground to roughly 9.5 lux, about the level of a street light. Under the beam, the paper says, the sky would look like dusk, with no stars visible. Outside it, the sky would still be brighter than under a full moon for about 14 kilometers, and the night would be altered for about 30 kilometers.

For comparison, the first-generation Iridium phone satellites, the fleet that lost Iridium 33 in the 2009 collision with Cosmos 2251, had flat, door-sized antennas that sometimes caught the Sun. These “Iridium flares” reached about magnitude -9.5 for a few seconds, and the satellites were all taken out of orbit by the end of 2019. The International Astronomical Union’s Centre for the Protection of the Dark and Quiet Sky asks operators to keep low-orbit satellites fainter than magnitude 7, too faint to see with the naked eye. As the dark-sky researcher John Barentine put it to Sky & Telescope, “the point is to be seen.”

Why astronomers are so alarmed

Astronomers are already living with a crowded sky. According to the American Institute of Physics, active satellites in low Earth orbit rose from about 1,200 to about 12,000 in a decade. Scientific American counts more than 14,500 active satellites, more than 10,000 of them Starlink. Those reflect sunlight by accident, streaking long telescope exposures. Reflect Orbital’s would do it on purpose.

The American Astronomical Society called the 50,000-mirror plan potentially “disastrous for professional and amateur astronomy, human health, and ecology”. It also warned of “potential for catastrophic interference with federally funded astronomical research”. Olivier Hainaut of the European Southern Observatory simulated the full constellation for Scientific American. He found the night sky worldwide could become three to four times brighter, and the darkest sites on Earth could become as bright as a suburban town. “We astronomers are certainly not in favor of forbidding satellites,” he said; the aim is “not to do anything stupid.”

Bakos was blunter in Sky & Telescope: “This is not a ‘concern’—this is like a bullet in your head… It’s the end of ground-based astronomy.” Samantha Lawler of the University of Regina said the FCC decision “really scares me”.

The company says it has thought about this. Its website promises “strict exclusion zones for astronomy” and says satellite positions will be shared in advance so that observatories can plan around them. A spokesperson told Scientific American, “We welcome oversight,” and the company told The Guardian it avoids astronomy sites and parks. Its website also says the light is not bright enough to harm eyes “even when viewed through a telescope”. However, AIP reported that the company’s own filing said telescopes of 12 inches or larger could cause eye damage if pointed straight at a mirror. The FCC called that a “small risk under unlikely circumstances”.

Birds, sleep and the right to darkness

A 2024 meta-analysis in Ecology Letters pooled 127 experiments on 31 wild vertebrate species. It found that artificial light at night cut melatonin secretion by an average of 36 percent. Melatonin helps set daily and seasonal rhythms. Many songbirds migrate at night, which is why environmental groups want a ban on illumination during peak bird migration.

Those groups are DarkSky International, Environment America, Public Employees for Environmental Responsibility (PEER) and the American Bird Conservancy, represented by Earthjustice. On August 6 they asked the full FCC to reverse the approval and require an environmental review under the National Environmental Policy Act. Failing that, they asked for 16 conditions, including a 160-kilometer buffer around observatories. “This is an idiotic project,” Earthjustice attorney Jan Hasselman told The Guardian. “It would really end the night sky as we know it.” DarkSky’s chief executive, Ruskin Hartley, called it “an unprecedented assault on the nighttime environment”.

David Ho, a climate scientist at the University of Hawaii, raised another worry with The Guardian: “Imagine wanting to turn those mirrors around so that the Earth gets hotter.”

Underneath all of it is the question Michelle Wooten, a co-chair of the AAS Committee for the Protection of Astronomy and the Space Environment, asked in Sky & Telescope: “Who owns the night sky?”

A gap in the rulebook

For now, nobody in Washington clearly regulates reflected light. The FCC licenses the satellite’s radio link. In its order it said optical concerns fell outside its authority, weighed the public-interest benefits and granted the license with conditions. AIP reported that about 1,800 comments were filed. The AAS filed the only valid petition to deny. Similar petitions from the Argentine and Spanish astronomy societies were rejected because they lacked the required certifications. On October 9 the FCC issued a limited order on reconsideration in response to a petition from Reflect Orbital itself. It did not grant the environmental groups’ request.

Other parts of the government are circling the problem. AIP reported that the Office of Space Commerce at NOAA is proposing a voluntary certification for novel space activities. Sky & Telescope also pointed out that under the Outer Space Treaty the launching state is liable for damage, a knotty question when American mirrors fly on Indian rockets. And every new satellite adds to the clutter that feeds the Kessler syndrome, and 50,000 large mirrors would be a lot of extra surface in orbit.

What happens next

One test satellite has a license, a launch contract and a target year. Fifty thousand is still a line on a roadmap, and each step toward it needs new licenses, money and customers willing to pay thousands of dollars an hour for light. “We’re pretty good at working at night already,” Cash told the Chronicle. The company argues that disaster zones and solar farms are real needs, and that its light can be aimed precisely and announced in advance.

Dark-sky advocates are organizing for a longer fight, with DarkSky’s Under One Sky conference set for November 6 and 7. Astronomy has always been built on faint light. The photographic plates that let Henrietta Leavitt find a yardstick for cosmic distances and the long exposures that helped Milton Humason chart the expanding universe both needed dark skies. Whether the sky above Mount Whitney stays dark may depend on regulators who have not yet agreed the problem is theirs.

If you enjoyed this, read how space junk can make more space junk, the other big worry about a crowded orbit.

Further reading