The return of the space mirror
Reflect Orbital, an aerospace company headquartered in Hawthorne, California, is preparing to launch a satellite designed to reflect sunlight onto the night side of Earth. Founded in 2021 by Ben Nowack and Tristan Semmelhack, the privately held firm aims to provide responsive lighting after dark and increase the effective hours of solar energy production. Nowack, who previously worked at Zipline and interned at SpaceX, serves as chief executive, while Semmelhack is the chief product officer. The company has attracted significant investment, raising $35.2 million in Series A funding from venture capital firms including Sequoia Capital, Lux Capital, and Climate Capital.
In July 2026, the Federal Communications Commission granted the company a license to launch its first demonstration satellite, Eärendil-1. The spacecraft is scheduled to fly aboard a SpaceX Falcon 9 rideshare mission later this year. Eärendil-1 is designed to operate at an altitude between 500 and 650 kilometres. The 134-kilogram satellite will unfurl four triangular panels to create an 18-by-18-metre square of highly reflective material, constructed with an aluminium front and a space-grade plastic backing. Once deployed, the satellite will use oversized reaction wheels (components typically used in 650-kilogram-class spacecraft) to angle its mirrors without burning fuel. The resulting illumination will cover an area about five kilometres wide and is expected to provide brightness comparable to a full moon. Because the satellite moves at approximately 7.5 kilometres per second, it will illuminate a single target on the ground for only a few minutes before passing over the horizon. The company received a $1.25 million award from the United States Air Force in 2025 to support this initial demonstration. Nowack has stated that the prototype cost around $2 million to build, but future production units are estimated to cost roughly $300,000 each.
The commercial case for artificial daylight
The primary market for this technology is utility-scale solar farms. Humanity consumes less than one trillionth of the sun's output, and solar energy currently represents a small fraction of global electricity production. Reflect Orbital intends to extend the productivity of existing solar infrastructure without requiring new land use or additional ground equipment. By 2030, the company aims to field enough satellites to supply solar farms with about 200 watts per square metre during the evening peak in electricity demand. This intensity is comparable to sunlight at dawn or dusk.
To achieve continuous illumination, a massive constellation is required. Reflect Orbital plans to launch 36 satellites by 2027, expanding to 5,000 by 2030, and ultimately operating up to 50,000 reflector satellites by 2035. Arranged in a continuous chain, these larger production satellites, which would measure 54 metres across, could keep solar farms illuminated for hours as new mirrors come into view just as previous ones set. The company also envisions selling its lighting services to emergency response crews, remote industrial sites, and outdoor events.
This architecture contrasts sharply with traditional space-based solar power, which involves collecting solar energy in orbit, converting it to microwaves, and beaming it to a ground rectenna. Such microwave systems face immense regulatory and technical hurdles. For instance, engineers frequently propose using the 5.8 GHz frequency for power transmission. However, 5.8 GHz is not regulatorily available for space-to-Earth, and securing spectrum rights remains a massive barrier. By merely redirecting natural sunlight, Reflect Orbital bypasses the need for complex orbital power conversion and ground rectennas entirely.
The historical precedent of Znamya
The concept of placing reflective surfaces in orbit to redirect sunlight to the ground predates Reflect Orbital by a century. The physicist Hermann Oberth first proposed large orbital mirrors in 1923. However, the only successful deployment of such a device was achieved by the Soviet Union and Russia during the 1990s under the Znamya project. Led by aerospace engineer Vladimir Syromyatnikov, who famously designed the docking system for the 1975 Apollo and Soyuz mission, the initiative originally sought to test solar sail propulsion but was repurposed to demonstrate artificial orbital illumination.
On 27 October 1992, the uncrewed Progress M-15 cargo spacecraft launched from the Baikonur Cosmodrome to resupply the Mir space station. After completing its primary logistical mission, the spacecraft undocked on 4 February 1993 and deployed Znamya 2. The device was a 20-metre-wide circular mirror made of aluminium-coated plastic that weighed less than four kilograms. Because the mirror lacked a rigid frame, the spacecraft had to spin to deploy the fragile material using centrifugal force. The experiment succeeded, reflecting a five-kilometre-wide beam of light that traversed Europe from southern France to western Russia at eight kilometres per second.
Despite this technical triumph, the project ultimately collapsed. Russia launched a larger successor, Znamya 2.5, in October 1998 aboard a Progress M-40 spacecraft. This 25-metre reflector was designed to produce a bright spot seven kilometres in diameter with a luminosity between five and ten full moons. During its deployment on 4 February 1999, the delicate mirror snagged on a spacecraft antenna and tore apart. After several failed attempts by Russian mission control to free the material, Znamya 2.5 was de-orbited and burned up upon re-entry, effectively ending the Russian space mirror programme.
The regulatory and environmental backlash
Reflect Orbital faces fierce opposition from astronomers, environmentalists, and dark sky advocates. The core argument against the constellation is that scattered light from thousands of orbital mirrors will significantly increase skyglow and degrade ground-based astronomical observations. Astronomer Miroslav Kocifaj and his research team calculated the illumination from these satellites and concluded that light pollution would extend well beyond the targeted areas. Critics warn that the proposed fleet could add several hundred objects as bright as Venus to the night sky by 2035. Moving beams of light from space-based mirrors could also create sudden flashes or glare, posing risks to night vision, aircraft operations, and wildlife ecosystems, including bird migration patterns.
Jonathan McDowell, an astrophysicist who has tracked human-made space objects since 1989, publicly criticised the initiative. In September 2026, he described the system to Science News as an insane project, arguing that the company fails to understand that darkness is a beneficial state. Furthermore, some astronomers assert that viewing the reflected beam through a moderate-sized telescope could damage observers' eyes or harm sensitive research equipment. Reflect Orbital counters that the risk of eye damage is extremely low, calculating that intentionally tracking the satellite for 100 seconds through a 12-inch diameter telescope would still stay within international eye-safety limits.
In March 2026, the American Astronomical Society petitioned the Federal Communications Commission to deny Reflect Orbital's request to operate its test satellite. When the agency granted approval in July, nonprofit organisations including DarkSky International formally asked the regulator to reverse its decision. The Federal Communications Commission concluded that it lacks the authority to evaluate the environmental impacts of an orbital solar reflector, limiting its review strictly to radiofrequency spectrum usage. DarkSky International continues to demand a full environmental risk evaluation by independent experts.
Reflect Orbital has defended its approach, stating that it engaged with the astronomical community before filing its licensing request. Christopher Buscombe, a spokesperson for the company, wrote in an email to Science News in September 2026 that "exploring how new technologies can help the world forgo fossil fuels is a responsible application of the scientific process." The company insists its service will be controllable, transparent, and avoidable. The reflectors are motorised and steerable, which the firm claims will ensure that reflected light is visible only in targeted areas and kept away from observatories and dark sky parks. Furthermore, the company plans to share its satellite positions in advance so that researchers can plan their observations around the brief windows of operation.
What would change my mind
The commercial viability of orbital mirrors depends entirely on achieving a massive scale while navigating an ambiguous regulatory environment. My current view is that Reflect Orbital offers a more immediate path to space-based solar power than microwave transmission, but several factors could alter this assessment.
First, the regulatory loophole that allowed Eärendil-1 to be licensed might close. If the Federal Communications Commission or international bodies assert jurisdiction over orbital light pollution, the company could be blocked from launching its planned 50,000-satellite constellation. Without the ability to deploy in vast numbers, the service cannot provide continuous illumination.
Second, the operational reliability of the constellation remains unproven. If the typical worst gap in a day between satellite passes is too long, utility-scale solar farms will not be able to rely on the service to meet evening peak demand. A failure to provide consistent, uninterrupted light would destroy the economic justification for the project.
Finally, the physical risks of operating massive, fragile structures in low Earth orbit cannot be ignored. The failure of Znamya 2.5 demonstrates how easily thin reflective membranes can snag and tear. Furthermore, as low Earth orbit becomes increasingly congested, the large surface area of these mirrors makes them highly vulnerable to space debris. A catastrophic collision could generate a debris field that endangers other assets, prompting immediate regulatory intervention that would ground the entire enterprise.
Comments